The new issue of *Psychological Bulletin* (vol. 134, #5) includes a
review article: “The Effectiveness of Psychotherapeutic Interventions
for Bereaved Persons: A Comprehensive Quantitative Review.”

Joseph M. Currier, Robert A. Neimeyer, and Jeffrey S. Berman are the authors.

Here’s how the article begins:

[begin excerpt]

Losing a loved one to death is an inherent part of human life. Reactions
to loss are as varied and multifaceted as the grievers themselves,
although research indicates that the majority of bereaved people tend to
experience strong emotions, a sense of cognitive disequilibrium, and
impaired role functioning for at least a short period (see Bonanno &
Kaltman, 2001, for a review). Despite the reality that almost every
person will experience the pain of bereavement at some point, and that
the majority will improve without formalized help, evidence suggests
that many persons do not improve naturally (Bonanno et al., 2002;
Bonanno, Wortman, & Nesse, 2004). For the 10% to 15% of bereaved persons
who experience intense suffering, sometimes for years (Ott, 2003), the
death of a loved one can precipitate a number of psychologically and
medically debilitating symptoms, which can even prove fatal (Prigerson
et al., 1997). Research on complicated or prolonged grief in particular
also demonstrates that the griever’s distress is often not reducible to
common psychological disorders (e.g., depression and posttraumatic
stress disorder [PTSD]), for which people receive other
psychotherapeutic interventions (see Lichtenthal, Cruess, & Prigerson,
2004, for a review). Moreover, recent studies have yielded evidence for
the relative efficacy of interventions designed for complicated grievers
as compared with more traditional treatments, such as interpersonal
(Shear, Frank, Houch, & Reynolds, 2005) and supportive psychotherapies
(Boelen, de Keijser, van den Hout, & van den Bout, 2007), thereby
highlighting the relevance of interventions sensitive to the needs of
those showing complicated grief symptoms and other clinically
significant indications of poor bereavement adaptation. The purpose of
this study is to provide a comprehensive quantitative summary of the
effectiveness of bereavement interventions over a range of therapeutic
outcomes for many different types of persons.

In general, years of outcome research leave little doubt that
psychotherapy provides an effective means of help for the majority of
distressed persons. Whatever theory drives the intervention and
regardless of whether the therapy targets an individual, a family, or a
group, studies usually conclude that treated clients are substantially
better off than their untreated counterparts (see Lambert & Ogles, 2004;
chap. 3 of Wampold, 2001, for reviews). However, attempts to evaluate
the effectiveness of psychotherapies for bereaved persons have not
yielded such conclusive results, and many important questions remain.
Some researchers have claimed that no strong empirical basis exists for
endorsing the effectiveness of these interventions (Jordan & Neimeyer,
2003; Schut & Stroebe, 2005); others have recently called for cautious
optimism and recommended that the field suspend judgment as to whether
bereavement interventions are less effective than other therapies
(Larson & Hoyt, 2007). Even though narrative reviews have pinpointed
studies that showed favorable outcomes, reviewers have reached different
conclusions regarding the state of the evidence for the absolute
efficacy of these interventions. These have included “no consistent
pattern of results for well-designed studies” (Forte, Hill, Pazder, &
Feudtner, 2004, p. 11), no empirical support for universal preventive
efforts combined with cautious support for interventions with more
distressed grievers (Schut, Stroebe, van den Bout, & Terheggen, 2001),
“a small amount of quantitative evidence” for universal approaches with
bereaved children (Curtis & Newman, 2001, p. 492), and mixed benefits
for bereaved spouses (Potocky, 1993) and families (Schneiderman,
Winders, Tallett, & Feldman, 1994).

Although researchers and clinicians increasingly have looked to meta-
analysis as a way to integrate studies on the effectiveness of
interventions, the existing quantitative reviews of bereavement
interventions have failed to reach a clear consensus (Allumbaugh & Hoyt,
1999; Currier, Holland, & Neimeyer, 2007; Kato & Mann, 1999). As part of
a larger qualitative review, Kato and Mann (1999) reviewed 11 studies
that randomly assigned bereaved participants to intervention and no-
intervention groups and examined modality and intervention type for
differences in outcome. In comparison with the encouraging effects
discussed in secondary reviews of outcomes in psychotherapy (Lambert &
Ogles, 2004; chap. 3 of Wampold, 2001), Kato and Mann found a meager
overall effect size (d = 0.11) and no clear moderators of outcome.
Unfortunately, these authors departed from conventional meta-analytic
procedures, thereby making it difficult to place full confidence in
their results. First, they likely violated the assumption of statistical
independence by treating different outcome measures within studies as
separate observations in the overall analysis. Second, they adopted an
overly conservative coding strategy and inferred effect sizes of d = 0
for four of the studies, each of which actually reported information
permitting statistical derivation of non-zero effect sizes. Third, on
the basis of their inclusion criteria, they failed to take into account
nearly two thirds (n = 19) of the randomized controlled trials available
for review before 1999.

[end excerpt]

Here’s how the Discussion section begins:

[begin excerpt]

The overall results from this review demonstrate that bereavement
interventions have a small but statistically significant effect
immediately following intervention but that therapeutic outcomes failed
to differ reliably from zero at later follow-up assessments. These
results contrast with meta-analytic reviews of general psychotherapy
demonstrating that treatments help substantially to ameliorate distress
symptoms and to improve functioning (Lambert & Ogles, 2004; Smith et
al., 1980; Wampold, 2001). Instead, the present results accord more
closely with the majority of smaller-scale qualitative (Schut et al.,
2001), quantitative (Currier et al., 2007; Kato & Mann, 1999), and
secondary (Jordan & Neimeyer, 2003; Schut & Stroebe, 2005) reviews, each
of which has concluded that, on average, recipients of bereavement
interventions are not appreciably less distressed when compared with
those who do not receive any formalized help. Beyond this general
conclusion, the present comprehensive review documents the relevance of
attending to the targeted population and reinforces the growing
consensus that psychotherapeutic interventions for bereaved persons can
be effective in instances when researchers and clinicians focus on
persons who are genuinely in need of help.

Contrary to treatment interventions in which clients meet some
predefined criteria for psychological disorder or present a specific
need for help, bereavement interventions are frequently practiced in
more of a preventive manner in which little attention is given to
requisite manifestations of distress. The present findings for grief
therapies are similar to those observed for interventions generically
applied after other types of traumatic experiences (van Emmerik,
Kampshuis, Hulsbosch, & Emmelkamp, 2002). Such evidence challenges the
common assumption in bereavement care that routine intervention should
be provided on a universal basis or according to special objective
circumstances surrounding the loss. Interventions targeting universal
populations failed to produce better outcomes than would be expected by
the passage of time, and although interventions with higher risk
grievers showed a benefit at posttreatment, the gains were relatively
small and the evidence failed to suggest that selective interventions
yielded statistically significant outcomes at follow-up. Conversely,
when the extra step was taken to assess for specific difficulties
adapting to loss as a requirement for treatment, effect sizes compared
favorably with the successes shown for psychotherapy in general. These
results converge with the conclusions of others in the field (e.g.,
Jordan & Neimeyer, 2003; Schut & Stroebe, 2005; Schut et al., 2001) by
pointing to an apparent relation between the level of bereavement-
related distress and the likelihood of achieving successful therapeutic
outcomes with bereaved clients. Viewed alongside the growing body of
evidence that the passage of time frequently does not alleviate
difficulties associated with maladaptive reactions to loss (Bonanno et
al., 2002, 2004; Lichtenthal et al., 2004; Ott, 2003), it is encouraging
that indicated grievers clearly were shown to benefit from intervention.

Aside from the targeted population and method of recruitment, no other
factors were shown to relate systematically to outcome. Several of these
analyses of moderators focused on person- and death-related risk factors
and features of the interventions themselves. The present evidence also
failed to show support for the timing of the intervention as a crucial
moderator of outcome. In view of prior quantitative reviews that
converged on a negative association between length of bereavement and
outcome (Allumbaugh & Hoyt, 1999; Currier et al., 2007), it was
surprising that even analyses of universal and selective samples did not
yield differential levels of effectiveness based on the average amount
of time elapsed since the loss. Insofar as it seemed possible that the
limited effectiveness of bereavement interventions could be attributed
to a reliance on aggressive outreach procedures (Larson & Hoyt, 2007;
Schut & Stroebe, 2005; Schut et al., 2001), such as soliciting bereaved
persons through obituaries or hospital records, it was also surprising
that self- and/or clinically referred participants were not shown to
achieve better outcomes across both assessments. Instead, the results
suggest that the method of recruitment was only critical at
posttreatment, at which time point studies intervening with referred
clients generated better outcomes than those relying on aggressive
outreach procedures. Although the current results highlight the need for
greater focus on the subjective aspects of adaptation to loss and the
identification of individuals presenting an immediate need of clinical
intervention, these discrepancies with previous work highlight the
importance of clarifying the issues of timing and recruitment with
respect to the delivery and study of interventions for bereaved persons.

[end excerpt]

Here’s the abstract: “Previous quantitative reviews of research on
psychotherapeutic interventions for bereaved persons have yielded
divergent findings and have not included many of the available
controlled outcome studies. This meta-analysis summarizes results from
61 controlled studies to offer a more comprehensive integration of this
literature. This review examined (a) the absolute effectiveness of
bereavement interventions immediately following intervention and at
follow-up assessments, (b) several of the clinically and theoretically
relevant moderators of outcome, and (c) change over time among
recipients of the interventions and individuals in no-intervention
control groups. Overall, analyses showed that interventions had a small
effect at posttreatment but no statistically significant benefit at
follow-up. However, interventions that exclusively targeted grievers
displaying marked difficulties adapting to loss had outcomes that
compare favorably with psychotherapies for other difficulties. Other
evidence suggested that the discouraging results for studies failing to
screen for indications of distress could be attributed to a tendency
among controls to improve naturally over time. The findings of the
review underscore the importance of attending to the targeted population
in the practice and study of psychotherapeutic interventions for
bereaved persons.”

The author note states that correspondence about the article may be sent
to Joseph M. Currier, Department of Psychology, University of Memphis,
202 Psychology Building, Memphis, Tennessee 38152; Email:
<> or to Robert A. Neimeyer, Department of
Psychology, University of Memphis, 202 Psychology Building, Memphis,
Tennessee 38152; Email: <>.