POSITIVE OUTCOMES FOLLOWING BEREAVEMENT : PATHS TO POSTTRAUMATIC GROWTH

Recent theory and research have drawn attention to the need to better understand the positive changes, termed posttraumatic growth, that often occur in bereaved individuals; even as negative emotions related to grief persist. We describe five dimensions of posttraumatic growth and present a model for understanding how the loss of a close other can eventually lead to a recognition of important positive personal changes. Loss, especially unexpected loss, disrupts an individual’s beliefs about the world and initiates a process of rebuilding an understanding. During this process, many people come to realise their own strengths, appreciate the impact of their relationships, and have new spiritual insights. A strategy for facilitating growth during clinical work also is described.

The idea that significant positive changes can occur in persons who are confronted with significant challenge, suffering, and loss is ancient (Tedeschi & Calhoun, 1995), and previous pioneering scholars of the 20th century also suggested this possibility (Caplan, 1964;Dohrenwend, 1978;Frankl, 1963).However, systematic investigations of this phenomenon are more recent and they suggest that many persons dealing with a wide range of major life stressors (Andrykowski, Brady, & Hunt, 1993;Elder & Clipp, 1989;Joseph, Wil-liams, & Yule, 1993), including bereavement (Calhoun & Tedeschi, 1989-90;Hogan, Greenfield, & Schmidt, 2001;Kessler, 1987), report significant positive changes from their struggle with their challenges and losses.
Several significant factors combined to encourage clinicians and researchers to begin to focus on growth per se.For example, Schaefer and Moos (1992) wrote a chapter on crisis and personal growth, O' Leary and Ickovics (1995) published a paper on "resilience and thriving in response to challenge", and we published the first book (Tedeschi & Calhoun, 1995) looking specifically at the phenomenon of positive change arising from the encounter with major life crises, such as bereavement.Quantitative ways of assessing growth also became available.For example, Joseph et al. (1993) described the development of the Changes in Outlook Questionnaire, Park et al. (1996) published their findings and introduced the Stress-Related Growth Scale, and we reported on the development of our own scale, the Posttraumatic Growth Inventory (Tedeschi & Calhoun, 1996).By mid-2009, a search using the PsychInfo system of the American Psychological Association listed 333 sources on "posttraumatic growth" and 71 on "stress-related growth." Although a variety of terms have been offered for this phenomenon, we will use the term posttraumatic growth (Tedeschi & Calhoun, 1995, 1996) to refer to positive changes experienced as the result of the struggle with major life crises.In this paper we will provide an overview of the general characteristics of posttraumatic growth, describe a model for how growth may unfold in the grief experience, with a particular focus on challenges to the individual's assumptive world (Parkes, 1971), and we will conclude with some suggestions for clinical work with bereaved persons.

Grief and posttraumatic growth
Available findings on the course of grief suggest that it does not unfold in neat stages, but it rather tends to be characterised by the oscillation between different psychological states and processes (Stroebe & Schut, 1999).For many people the distress triggered by loss does not dissipate within a few short months, but it can last for many years (Carnelley, Wortman, Bolger, & Burke, 2006;Rogers, Floyd, Seltzer, Greenberg, & Hong, 2008).Conversely, however, the assumption that the loss of a loved one inevitably produces high levels of enduring psychological distress also has been found to be incorrect, at least for a significant proportion of bereaved persons (Bonanno et al., 2004).There is also evidence that for some people, perhaps many, coping with bereavement can provide the context for significant positive changes, i.e., posttraumatic growth (Cadell, Regehr, & Hemsworth, 2003;Engelkemeyer & Marwit, 2008;Matthews & Servaty-Seib, 2007;Wagner, Kanevelsrud, & Maercker, 2007).
Although the observation that the grief experience can include the possibility of personal growth is also not new (Nadeau, 2008), many investigations of the grief process have failed to attend to this phenomenon (Stroebe, Hansson, Schut, & Stroebe, 2008b).The available research suggests that posttraumatic growth is not uncommon in people facing major difficulties, including bereavement.
In this paper, we will focus the discussion on growth in the context of grief; changes that persons regard as significantly positive and that emerge from their struggle with loss.Before proceeding further, however, it is important to keep several things in mind.Posttraumatic growth clearly occurs in a context of significant life challenges, with concomitant states of psychological distress and sometimes great suffering.It would be a gross misinterpretation of what we are saying to assume that the focus on growth minimises the importance of attending to the negative responses that for many can accompany the experience of loss.Further, we tend to regard the assumption that the experience of growth will, or should, produce a commensurate reduction in psychological distress as erroneous.The data are somewhat inconsistent on the relationships between various measures of growth and general measures of psychological distress (Helgeson, Reynolds, & Tomich, 2006), but the available evidence does suggest that the experience of growth is best viewed as statistically independent from the experience of posttraumatic losses (Baker, Kelly, Calhoun, Cann, & Tedeschi, 2008;Cann, Calhoun, Tedeschi, & Solomon, in press).It is also important to remember that not all persons confronted with major stressors, including bereavement, experience growth.Another reminder is that we are not suggesting that the encounter with loss and trauma is the only context in which individuals can grow and develop; the focus here is on the process of growth that does occur in circumstances that have historically been viewed, especially among behavioural scientists of the 20th century, as tending to produce only negative psychological consequences.With these caveats in mind, we turn now to the general domains of posttraumatic growth.
Posttraumatic growth tends to be reflected in changes experienced in five different areas (Morris, Shakespeare-Finch, Rieck, & Newberry, 2005;Taku, Cann, Calhoun, & Tedeschi, 2008), and one of these is self-perception.The change in self-concept that reflects growth can be summarised with the somewhat paradoxical phrase more vulnerable, yet stronger.The loss of a loved one, particularly when the death is violent or sudden, tells the survivors that they are indeed vulnerable to losses that are unpredictable, unexpected, and perhaps tragic.The experience of loss, however, can also lead many persons to experience themselves as stronger and more self-confident (Calhoun & Tedeschi, 1989-90;Carnelley et al., 2006).As one bereaved parent told us: "I've been through the absolute worst that I know.And no matter what happens, I'll be able to deal with it."This brief statement reflects the view that the person has indeed been, or may still be, experiencing significant psychological distress and major challenges to adjustment and adaptation.But, along with the difficult and negative aspects, there is the assumption that simply surviving "the worst that I know" offers the person evidence of their personal strength in the context of adversity (Linley & Joseph, 2004).
Another domain of posttraumatic growth is an experience of changed relationships with other persons.Certainly, as with any domain of growth, crises and losses can produce negative changes in relationships; but many bereaved persons also describe positive changes in their relationships with others.One of the items of the Posttraumatic Growth Inventory reflects this dimensionthe experience of an increased sense of closeness with others.This closeness is often expressed about significant personal others, such as members of one's family and close friends.However, there is also the indication that this sense of connectedness to others is experienced as a feeling of greater compassion toward others in general, or to others who share similar difficult losses.As one bereaved person said: "I've become very empathic towards anybody in pain and anybody in any kind of grief." When people die, the bereaved may need to take over responsibilities and relationship connections that were formerly ascribed to the deceased.This seems to be especially the case in spousal bereavement (Calhoun & Tedeschi, 1989-90).Although initially these new roles can be burdensome, they can often open the bereaved to a third domain of growth, new possibilities they had little or no experience with before.Loss also can open up the possibility of new relationships.Of course the person who has died cannot be directly replaced, and bereaved persons seldom wish to think this way, but the vacancy produced in the bereaved person's life allows for new people to enter.
When a death occurs, people can report that there is a realisation that the end of life may be sooner than they think, or that they must make the most of what time they have.This appreciation of life, living it more vividly, may be difficult for some people to sustain, but it is sometimes consolidated into new habits of living more deliberately rather than routinely.This is a fourth domain of posttraumatic growth.
The fifth general area of posttraumatic growth includes existential elements, and for many persons, depending on particular geographic and cultural contexts, it can also reflect spiritual and religious elements.The metaphor of the "thin places" of Celtic mythology (locations where the borders between this world and 'the other' are more permeable), or the more existential idea of mortality salience (Martin, Campbell, & Henry, 2004), offers a way of thinking about this dimension of posttraumatic growth.Bereaved persons, because of their experience with the death of a close other, may experience changes in the way they understand themselves, their existence as mortal human beings, and, for some, their connection to something transcendent.This dimension of growth has been reported in several studies of the bereaved (Balk, 1999;Rosenblatt, 2000).Grieving persons may experience posttraumatic growth as a result of existential questioning, irrespective of specific spiritual or religious beliefs.For many, however, particularly in the United States, such existential growth includes spiritual or religious elements (but this may not be the case in places where organised religion is less important, such as Europe and Australia).
Religion/spirituality may serve as both a framework promoting growth through increased sense of meaning and purpose (Rogers et al., 2008), as well as a form of growth through deepening faith and conviction.Finding meaning in one's life following a loved one's death can represent a spiritual change that may result from bereavement (Balk, 1999).The often difficult experience of losing a loved can challenge the belief that events in one's life have meaning and this disruption may initiate a process of reexamination of important components of the assumptive world (Batten & Oltjenbruns, 1999).
In a group of bereaved parents, for example, spiritual coping was among the top three predictors of posttraumatic growth (Znoj, 2006), suggesting a possible avenue towards growth.Spirituality has also been found to be an important source of growth for bereaved caregivers of persons with HIV (Cadell, 2007).Continued connections to the deceased can, for some persons, encourage increased spirituality or a desire to maintain spiritual beliefs (Cait, 2004;Sormanti & August, 1997); with the deepening of spirituality representing spiritual growth.A caveat, however, is that the continued connection with the deceased may or may not be generally adaptive (Neimeyer, Baldwin, & Gillies, 2006;Parker, 2005;Stroebe & Schut, 2005).
These five domains clearly do not include every single possible way in which grieving persons may experience growth.Particular circumstances may lead individuals to describe stressor specific changes (e.g., my father died of a heart attack and as a result I try to eat a much healthier diet) or idiosyncratic changes (e.g., the loss led me to want to go into nursing).However, the five broad domains reflected in the items of the Posttraumatic Growth Inventory do seem to capture a wide range of changes common to many circumstances, including bereavement.

A model of growth in the context of grief
Our earlier theoretical model of posttraumatic growth (Tedeschi & Calhoun, 1995) has been refined over time, based on empirical tests of components of the model.A current version of the model, outlining the processes through which highly stressful circumstances, such as the death of someone very close, can lead to the experience of posttraumatic growth is presented in Figure 1.In what follows, we will provide a brief overview of the elements of the model that are most relevant to bereavement.We will also consider evidence in support of the assumed processes.More extensive descriptions of our complete model of growth, including elements that we do not describe here, can be found elsewhere (Calhoun & Tedeschi, 2004, 2006;Tedeschi & Calhoun, 2004a).
Figure 1 A model of growth in grief Before they experience the death of a close other, individuals will differ in a variety of personality traits and in the many and varying components of their assumptive worlds.Assumptive world beliefs have been considered a key factor in determining the potential for posttraumatic growth by several theorists (Beder, 2004(Beder, -2005;;Gillies & Neimeyer, 2006;Janoff-Bulman, 1992, 2006;Tedeschi & Calhoun, 1995).Under typical circumstances, this broad set of beliefs helps individuals maintain a sense of how events in the world should unfold, and how they can influence events.They allow individuals to plan and experience the world as predictable, understandable, and meaningful (Parkes, 1971).Differences in assumptive world beliefs can exist based on cultural differences that encourage varying world views (Calhoun, Cann, & Tedeschi, in press;Nisbett, 2003) or based on previous life experiences that have shaped particular assumptions.
The death of a close other usually brings with it the experience of emotional distress.Regardless of the circumstances of the death, some sense of loss and distress is likely.The longer term impact of the death may depend on other, related factors.For example, Bonanno (2004) suggests that differences in resilience are related to the long term adjustment issues that often covary with bereavement.Resilient individuals experience the initial distress, but recover with little need for grief work, while those who are less resilient may continue to experience distress and benefit from working through their grief.
Differences in resilience may represent factors separate from variations in assumptive world beliefs, but the noted differences in responses to a death also should be related to differences in individuals' assumptive world beliefs.Models of posttraumatic growth focus on the critical importance of the degree to which the death, or any significant traumatic experience, challenges or shatters world assumptions (Calhoun & Tedeschi, 2006;Janoff-Bulman, 1992, 2006;Tedeschi & Calhoun, 1995).When an event challenges the existing world beliefs, the individual must deal not only with the distress resulting from the death, but the guiding principles about how the world is assumed to work also may need to be reassessed.Although the challenge to world beliefs sets the stage for possible growth, as individuals more closely examine the world and their place within it in light of the traumatic event, the disruption also can initially add to the distress experienced.
Evidence in support of the importance of considering the individual's world views can be drawn from the finding that some deaths present much greater psychological challenges than others.A "natural death", at the end of a long life, tends to be much easier to accept (Lehman, Wortman, & Williams, 1987), and allows for a clearer sense of meaning to be found (Gillies & Neimeyer, 2006).On the other hand, unexpected deaths that are less consistent with most assumptive world views, such as the death of a young child, are less "natural", and often lead to greater distress and more growth (Tedeschi & Calhoun, 2006).In the model in Figure 1, distress is expected regardless of the challenge to the assumptive beliefs, but for those for whom the death is consistent with their assumptive worlds, there is little other cognitive work required.They deal with their loss, survive the distress, and return to a stage of well-being much like the individuals Bonanno (2004) labels resilient.Alternatively, those whose beliefs have been seriously challenged cannot as easily move on.The death may have shattered assumptions, and this disruption requires attention to the rebuilding effort at the same time the distress of the death is being experienced.
Despite the essential importance of the challenge to the assumptive world beliefs in models of posttraumatic growth, until recently no effort had been made to assess directly differences in the disruption of assumptive beliefs.Janoff-Bulman (1989) developed a measure of global assumptive world beliefs, the World Assumptions Scale, and showed differences in these beliefs exist between people who had and who had not experienced traumatic events.However, no research had tried to directly assess the degree to which a specific event had led participants to engage in an examination of their core beliefs about the world.The Core Beliefs Inventory (CBI: Cann, Calhoun, Tedeschi, Kilmer, Gil-Rivas, Vishnevsky et al., 2010) was developed for this purpose.
On the CBI, participants rate the degree to which a recent highly stressful event led them to reexamine a number of core assumptions about themselves and their world.Scores on the CBI have consistently been shown to be predictive of levels of posttraumatic growth, both in analyses of cross-sectional and of longitudinal data, involving a variety of stressful events; greater disruption is associated with greater growth (Cann et al., 2010).CBI scores also were positively related to the level of distress experienced at the time of the event and to symptoms of posttraumatic stress.However, a global measure of distress was not predictive of later posttraumatic growth when CBI scores were included in the prediction models (Cann et al., 2010).
Thus, to the extent that a death disrupts core elements of the assumptive world, individuals experience more distress, show higher symptoms of posttraumatic stress disorder, and also have the potential for greater posttraumatic growth -it is the level of disruption of core beliefs which best predicts growth.Deaths which do not lead to significant reexamination of core beliefs should result in less reported growth.Future researchers might want to be cognisant of the potential importance of determining the degree to which individuals experience a death as an understandable and comprehensible event, consistent with their assumptive world beliefs, or see the death as "unnatural", and therefore as a challenge to their assumptive worlds.
The cognitive work associated with attempts to rebuild challenged or shattered assumptions begins as the person is also coping with the emotional distress of the unexpected or unnatural death of a close other (Calhoun & Tedeschi, 2006;Tedeschi & Calhoun, 2004a, 2004b).The model assumes that rumination plays an important part in the attempts to repair or reconstruct a workable belief system in the aftermath of the death.Earlier views of rumination associated such cognitive activities with increased distress and depression (Nolen-Hoeksema, McBride, & Larson, 1997;Segerstrom, Tsao, Alden, & Craske, 2000).However, more recent considerations recognise the multiple forms and impacts of rumination (Segerstrom, Stanton, Alden, & Shortridge, 2003;Treynor, Gonzalez, & Nolen-Hoeksema, 2003;Watkins, 2008).Rumination can be constructive or unconstructive, depending on whether the rumination supports continued negative thoughts and emotions or helps move the person toward problem solving or finding meaning.In our work, and in the model in Figure 1, we have distinguished between intrusive ruminations, the types of thoughts that are less controlled by the individual and are almost always unwanted and are associated with PTSD symptoms, from those ruminations that are more deliberate, focused on understanding a challenge and rebuilding a functional world view.
We assume that, due to the impact of the distress caused by the death and the challenged assumptions, the individual will commonly experience unwanted, negative, intrusive thoughts in the immediate aftermath of the death.As the distress is managed and the person is able to engage in more goaldirected ruminations, deliberate, constructive thoughts will tend to predominate.Although the model suggests a movement from one style of rumination to the next, it is more likely that both styles coexist, or may oscillate (Stroebe & Schut, 1999), as the intrusive thoughts serve as a stimulus to work even harder to engage the deliberate thoughts.Ideally, the frequency and disruptive impact of the intrusive thoughts will decrease as the deliberate rumination provides a reconstructed world view and allows the individual to find meaning in the death experience.However, a failure to successfully rebuild functional assumptive world beliefs could be associated with continued high levels of intrusive rumination, and, potentially, continued distress.
During this stage of cognitive reconstruction, the individual's efforts can be aided by socio-cultural influences which can guide the thoughts in a more constructive direction, facilitating growth, or, potentially, in a less constructive direction.For example, it has been shown that having a model, knowing someone who has had a similar experience and who has experienced growth, facilitates growth (Cobb, Tedeschi, Calhoun, & Cann, 2006;Weiss, 2004).Evidence also exists for broader influences through cultural rituals that can provide a context for understanding and dealing with events like death (Bonanno, Papa, Lalande, Zhang, & Noll, 2005).Of course, not all culturallybased beliefs or rituals are helpful to the process of grieving.For example, when people who commit suicide are not accorded the same degree of respect as people who have died from other causes, the bereaved family might experience a version of disenfranchised grief (Doka, 1999;2008).
A number of recent studies, in which different styles of rumination have been assessed, support the assumed processes in which deliberate rumination will be more strongly predictive of eventual posttraumatic growth, while intrusive rumination early, soon after the death, but not recently, will be associated with posttraumatic growth.Taku, Calhoun, Cann, and Tedeschi (2008), in a sample of bereaved Japanese university students, found that recent intrusive rumination was associated with greater distress, while deliberate rumination soon after the event predicted greater growth.Similarly, in a comparison of samples from the United States and Japan, it was found that across both samples, intrusive rumination soon after the stressful event was positively associated with growth, as was recent deliberate rumination (Taku, Cann, Tedeschi, & Calhoun, 2009).
Finally, in a study that assessed both positive changes, posttraumatic growth, and negative changes in the same domains as growth, posttraumatic depreciation, recent deliberate rumination was positively associated with growth, while recent intrusive rumination was negatively associated with growth.In predicting posttraumatic depreciation, however, recent intrusive rumination was a positive predictor, indicating a negative role for persistent intrusive rumination (Cann et al., in press).Taken together, these studies indicate that deliberate rumination is consistently associated with eventual posttraumatic growth; however, intrusive ruminations are associated with growth when they occur early, soon after the event, and do not persist unabated.Recent intrusive ruminations, perhaps indicative of a failed attempt to restore the assumptive world beliefs, are associated with less posttraumatic growth and greater posttraumatic depreciation.
Note that in the model, eventual success in rebuilding the beliefs and assumptions that comprise the assumptive world is believed to be associated with growth, and, perhaps, with a more complex appreciation of the world and other human beings.Simplistic notions of fairness and control over unpleasant experiences, for example, may be replaced with a more realistic recognition that bad things can happen even when people are good.This new world view may not always be associated with higher levels of well-being.People may have come to recognise their own strengths, to feel closer to others, and to have a new appreciation for life or life direction, but they may also now accept that the world is not always a fair or controllable place, and that bad outcomes cannot always be prevented.
In looking at the data we have collected over a number of years, across multiple studies, we note how bereavement, as a specific experience, differs from other highly stressful events in terms of eventual posttraumatic growth.In an accumulated sample of over 800 participants from the United States, who have completed both the Posttraumatic Growth Inventory (PTGI) and the Core Be-liefs Inventory, 233 participants reported on the death of a close other.Within this bereaved subgroup, scores on the CBI were positively correlated with all five subscales of posttraumatic growth (all r's > .43,all p's < .001),indicating that for those bereaved persons, disruption of core assumptive world beliefs is associated with eventual posttraumatic growth.A comparison of those reporting on a death (n = 233) with those dealing with other events (n = 571), indicates that the two groups do not differ on total posttraumatic growth scores or CBI scores (t's < 1), but differences are evident on the subscales of the PTGI.Those reporting on a death showed reliably more growth in the areas of Relationships with Others, Appreciation of Life, and Spiritual Change.However, bereaved persons reported less growth in the areas of Personal Strengths and New Possibilities.Although these data represent reactions of individuals dealing with a wide variety of deaths, some probably more "natural" than others, they do provide some hints about the growth experiences of individuals facing bereavement and grief, at least in a North American context.

Posttraumatic growth and clinical practice with grieving persons
There has been a good deal of controversy about the usefulness of clinical interventions with grieving persons.Some data have tended to show that grief therapy is quite unnecessary for most bereaved persons, but for those who do seek professional help, there are some outcomes that are robust and comparable to other forms of psychotherapy (Bonanno & Lilienfeld, 2008;Hoyt & Larson, 2008).Therefore, it is important to consider the persons for whom grief therapy is designed.
In our general model of posttraumatic growth we have emphasised that trauma is, to a great extent, defined by the degree of challenge to the assumptive world, and that posttraumatic growth develops, to a significant but not exclusive degree, out of an attempt to come to new understandings of a world that no longer fits people's ideas about themselves, how others behave, what their future will be, and the like.Some bereavement circumstances are more likely to challenge these assumptions than others, and some assumptive worlds may be more vulnerable to these challenges than others.Bonanno (2004) has shown that a significant proportion of people are not prone to these difficulties, i.e., they are resilient in the face of bereavement.On the other hand, people who demonstrate posttraumatic growth are likely to be those whose grief process includes challenges to their assumptive worlds, including their sense of meaning and purpose, or their understanding of themselves (Davis, 2008).
Just as with other difficult experiences, where challenges to core beliefs, high levels of distress, and rumination are associated with posttraumatic growth, the same appears to be the case with bereavement.When threat to self and a need for self-understanding is part of the grief process, posttraumatic growth is a result (Balk, 1999;Davis, Wohl, & Verberg, 2007).Grief therapies have been described that fit with this understanding of what is needed most by persons who are truly in need of intervention: a reconstruction of their beliefs, sense of meaning, and life narrative (Leighton, 2008;Malkinson, 2007;Neimeyer & Wogrin, 2008).
We have described our own framework for intervention, based on our work with posttraumatic growth (Calhoun & Tedeschi, 1999), as an approach that has relevance to work with persons who have experienced a variety of life crises that make necessary a reconstruction of beliefs, meanings, and the life narrative.This model of expert companionship is certainly appropriate for work with bereaved persons, and grows out of significant clinical experience with them (Tedeschi & Calhoun, 2004b;2006).In working with bereaved parents in particular, we have found a number of concerns that set people on a course that can include posttraumatic growth.There is a high level of distress, a shattering of the expectation that one will outlive one's children, disappointments and surprises about who is, and who is not, understanding and supportive, and questions about the afterlife, all of which are challenges to core beliefs.
This approach that attends to the possibility of posttraumatic growth is not a new form of therapy, but a particular perspective, and it fits well with cognitive, humanistic-existential, and narrative-constructionist approaches.There are several aspects to the posttraumatic growth therapy we call expert companionship that are especially relevant for bereaved persons.In the following we provide very brief descriptions of some elements of this particular clinical stance.

Humility & respect, not platitudes
A basic respect for the beliefs and experiences of bereaved persons, often rooted in their culturally-based understandings of death and grief, is essential to setting the stage for posttraumatic growth.Instead of seeking to merely provide comfort and reassurance with platitudes, that are often given by wellmeaning friends and family, the clinician working as an expert companion is willing to explore these beliefs, and the doubts about them, that may be raised by the experiences of the bereaved.

Constancy
The expert companion is willing to tolerate the fact that bereavement can be a longer-term process than what is expected by friends and family, or the bereaved persons themselves.The clinician is a constant in spite of oscillations (Stroebe & Schut, 1999) in the grief experience of the individual.

Tolerance of the strange, non-rational, and ambiguous
Experiences of death and bereavement can be strange and mysterious for some.People in these circumstances can be reluctant to talk about such things as belief in paranormal phenomena, continuing bonds (Klass & Walter, 2001), or unpredictable bouts of distress.The expert companion is constant as a non-judgmental listener to all these experiences.

Courage to hear
The expert companion may also need to listen to unpleasant stories of death, when illness, accident, or personal violence has produced gruesome and traumatic images of death that the bereaved persons may carry with them.

Appreciation of paradox
In order to facilitate posttraumatic growth, the expert companion must be able to appreciate the two sides, or more, of the stories told by the bereaved.For example: in the vulnerability of grief emerges strength; from doubtful questioning, new insights can emerge; and in the need for support, greater independence can be achieved.
Clinicians working with grieving persons must remember that the process by which posttraumatic growth may unfold occurs in the process of grief itself.A fundamental concern is the timing of discussions of growth, and the attributions made for the cause of posttraumatic growth.The expert companion is sensitive to the readiness of people to consider emerging indications of growth, and how these have come about.Some bereaved persons may decide at the outset that they are going to "make something good come of this", but the vast majority are simply trying to get through their distress in the early phase of their loss.When clinically appropriate (Zeldow, 2009), the expert companion can bring to the attention of the bereaved person indications of change in the five domains of posttraumatic growth as they seem ready to engage them more deliberately.The clinician should take care to attribute these changes to the struggle with the grief and loss, not to the loss itself.Bereaved persons are, of course, reluctant to think that the loss, e.g., the death of a husband, has produced something good per se.They are more willing to see that their struggle has produced something positive.It is not the loss itself, but the cognitive and emotional work, and the reconstructed assumptive world, including the life narrative, that produces change.This matter also highlights the crucial difference between what has been termed "perceived benefits" and posttraumatic growth (Davis, 2008).Benefits such as inheritances can certainly come in the aftermath of loss, but they are different in quality from personal growth, and are more likely to be accepted with ambivalence.

A possible framework
In order to attend to, and when appropriate encourage, the process of posttraumatic growth, clinicians may find our model of posttraumatic growth a useful framework when working with grieving patients.The model suggests where in the process the bereaved person may be, and what kind of therapeutic work might be most appropriate.In the early stages, emotional distress will need to be addressed, but not smothered entirely.Some degree of distress may be useful for producing posttraumatic growth.Rumination that is intrusive and unconstructive can be encouraged to become more deliberate, and focused on questions that are indicative of the development of posttraumatic growth, e.g., changes in role and identity, new ways of relating, existential or spiritual beliefs, and the purpose and meaning of life after the experience of loss.
Although posttraumatic growth may be very much an internal experience, with changes in views of self and the world that are not always apparent to the observer, there are ways in which posttraumatic growth can be enacted in life choices, relationships, and priorities.The expert companion can encourage, when appropriate, the enactment of posttraumatic growth so that the changes are not lost over time, but become a valued memorial to the person who has died.Through these changes, the adaptive bond with the deceased can be continued, and in this way, the suffering of bereavement is made meaningful.This does not mean that distress is eliminated, but that it is mitigated by the recognition that the loss has set in motion a reconsideration of life that has wrought valuable changes.