My early career in medical social work confronted me with the losses and consequent grief that are part of the life-course, in particular the specific losses of illness/disability, and death and dying. This led me to paying a growing attention to loss and the nature of grief. I was interested to begin my engagement with loss and grief by specifically focusing on bereavement. I had the opportunity to do this in a study for the Diocese of Lichfield exploring bereaved people’s experience of grief (Machin, 1980). This study was then incorporated into my Master’s degree project “Bereavement: an experience of loss,” building on the evidence of participants’ experiences to investigate their support needs and how best to meet those (Machin, 1985). My studies took me into a world dominated by the (then) current literature on grief and bereavement (i.e., Kubler-Ross, 1969; Bowlby, 1980; Parkes, 1972). Their writing was setting out the idea of stages of grief and the characteristics associated with various forms of loss; death and dying, separation in relationships, and bereavement. It was in the context of this background that my project took place. Participants in my study (n = 97) came from a variety of socio-demographic backgrounds. I used a structured interview schedule to explore the chronology of their bereavement experience. Alongside this I captured their qualitative reflections which were triggered by the interview questions. The findings emphasized the wide-ranging diversity in grief reactions. In an attempt to take wider learning from the research, I was interested in looking at the data to see if, despite the diversity, there were indications of a pattern of loss response which spanned those individual differences in grieving. While the qualitative narratives contained all the personal nuances of a grief experience, I believed a discernible pattern was evident. This was captured in three words often used by respondents to describe the impact of loss; first, a state of being “overwhelmed” and second, a predisposition to remain in “control.” What is represented by these two words seemed to suggest opposite positions on a spectrum of expression/suppression of feelings, thinking and behavior, which capture an element of the diversity of grief; diversity spanning overt and covert modes of reaction to loss. The third word used was “balance,” expressing the need to restore some equilibrium in the disturbances and disruption of grief. This was the first theoretical conceptualization of what was to become the RRL, as the three associated concepts, began to emerge.

At this stage, in identifying a pattern consistent with the idea of a “range of response to loss,” the work on “attachment” style and its link with loss (Ainsworth et al., 1978; Bowlby, 1980) seemed to have parallels with my proposition: identifying anxious/ambivalent attachment with the concept of “overwhelmed,” avoidant attachment with the concept of “control” and secure attachment with the notion of “balance.”

The Master’s research also identified the lack of available support for 44.3% of bereaved respondents. This became the incentive to pilot a project exploring the feasibility of a service for bereaved people in North Staffordshire. I led the pilot which successfully demonstrated a need and resulted in the foundation of a service—“Bereavement Care.” The service has continued to grow and is still operating, now known as the “Dove Service.” In 1984 I became its first Director and also had roles as a counselor/practitioner, supervisor and trainer. In this practice context I was hearing and observing experiences and expressions of grief that echoed the patterns of “the range of response to loss” which I had begun to identify.

In 1990, while still engaged in counseling practice, I took up the post of Lecturer in Social Work and Counseling at Keele University. At this time, I emersed myself in the study of (then) contemporary theoretical perspectives on loss and grief, alongside becoming engaged with a range of organizations and practitioners working in the field of bereavement in the UK; leading to a publication on guidance for practitioners (Machin, 1990/1998). I went on to work alongside a small group of academics and practitioners whose focus was on bereavement research, and together we established the “Bereavement Research Forum” supported by Dr. Colin Murray Parkes (a psychiatrist, researcher and leading expert on grief and bereavement support). During this period, I was presenting my emerging ideas on the “range of response to loss” to a variety of audiences so as to widen the discussions on the concepts and the practice implications. My experience of teaching, research, practice, and the conversations generated through this activity, further reinforced the emergent pattern of a “range of response to loss.” This pointed to a pattern consisting of the two contrasting grieving reactions, overwhelmed and controlled, and their implications for achieving a sense of balance (see Figure 2).

While I had seen parallels between Attachment Theory (Ainsworth et al., 1978; Bowlby, 1980) and the patterns I was noting, further developments were continuing at this period (e.g., Ptacek and Gross, 1997; Simpson and Rholes, 1998). Even within the post-modern research climate in which many earlier theoretical assumptions were being challenged and revised (Wortman and Silver, 1989; Stroebe, 1993), as a theory attachment clearly remained penetrating in its detail and robust in its empirical scrutiny. Emerging literature and research (Bartholomew and Shaver, 1998; Mikulincer and Florian, 1998; Simpson and Rholes, 1998) was taking concepts of “attachment” further. Mikulincer and Florian (1998) looked at the emotional and cognitive reactions to stressful events and observed three patterns. The first was associated with anxious/ambivalent responses where people see stressful events as threatening, irreversible and uncontrollable; giving a fuller explanation of the overwhelmed state identified in the “range of response to loss” pattern. Secondly, people who showed avoidant responses and dealt with stressful events by restricting their acknowledgment of distress and by being compulsively self-reliant; identified in the “range of response to loss” concept of control. The third was associated with people who gave evidence of being securely attached, and who were resilient in the face of adversity and whose constructive attitude toward life is a buffer against psychological distress; could be seen as a much deeper description of what it might mean to be “balanced.” This further articulation of Attachment Theory detailed the nature of the characteristics of loss more fully, providing greater depth for the RRL concepts.

Challenging the stage-based paradigm emanating from Freud’s (1917) work, Stroebe (1993) and her colleagues (Stroebe and Schut, 1999) devised an alternative model of grief: the Dual Process Model. This model outlines a process of oscillation between rumination on painful emotions and attention to social adaptation to loss; a movement between loss orientation and restoration orientation. This new theoretical perspective could also be seen to have evident parallels with the “range of response” pattern I was observing: loss orientation with overwhelmed; restoration orientation with control; and oscillation with balance.

Horowitz et al. (1979) had identified intrusion and avoidance as key elements in the face of challenging or traumatic events, and used them as the basis for the Impact of Events scale. These characteristics can also be seen as having characteristic parallels with the notion of overwhelmed and controlled grief reactions I was observing.

The third component of the “range of response to loss” pattern I originally defined as “balance,” i.e., the capacity to make adjustments to competing inner feelings and thoughts and the external challenges demanding adjustment to change. With encouragement from Dr Colin Murray Parkes (see reference above to his role in the research community), I re-defined the component as “resilience.” This brought further exploration of the unique contribution to understanding the capacity for human resilience, in the face of loss, from holocaust survivors (Valent, 1998) and from the growth of positive psychology (Seligman, 1999). Seligman moved away from the medical model’s attention to pathological and abnormal manifestations of grief, earlier defined by him as learned helplessness, by turning attention to the characteristics of resilience or learned optimism: acceptance of one’s own and other people’s feelings, even if they are expressed differently; ability to confront the reality of loss experiences and their consequences; ability to engage with possibilities and choices despite the limitations generated by loss; willingness to be open in the offering and receiving of support; having an overall sense of hopefulness; a capacity to make sense of life losses.

The evolution of traditional theories on loss and grief and the emergence of new ones provided an important base from which to test and give further definition to the “range of response” pattern.

Despite the evident parallels with other theories, it was important to test the propositions in the identified “range of response to loss” pattern and to validate the distinctions made between the three categories, i.e., overwhelmed, controlled and resilient. I did this through my doctoral research which took place at Keele University, with field work located in Bereavement Care (see above). As part of this process, I devised the Adult Attitude to Grief scale (AAG), specifically designed to capture the three “range of response to loss” concepts. The rationale for devising an attitude scale was based on the work of Bowlby (1980) who had set out the concept of an “internal working model” suggesting that individual cognitive, emotional and behavioral perspectives are developed over time, see Holmes (1993). Similarly, Parkes (1996) proposed an “assumptive world” view embodying attitudes that shape how life circumstances are navigated. The AAG is a self-report measure consisting of nine-items scored on a five-point Likert scale from “strongly agree” to “strongly disagree.” The items aim to elicit a life view and a self-view, in respect of a current loss. Table 1 shows the relatedness of the “range of response to loss” concepts to Attachment Theory and resilience concepts, and their specific characteristics which were represented in the AAG scale.

Ninety-four people took part in the study, of whom 64 agreed to a repeat interview 6 months later to capture reflections on using the AAG. In addition to the AAG, three standardized measures were used to test the validity of the concepts; Impact of Events scale (Horowitz et al., 1979), the Beck Depression Inventory (Beck et al., 1961) and the Leiden “Detachment” scale (Cleiren, 1991). The questionnaires were administered by practitioners working in the service.

Factor analysis confirmed the validity of the three groups of three statements, associated conceptually with the categories which now could be seen as a “framework” for understanding the range of response to loss (RRL) (Machin, 2001). In addition to the statistical evidence, given the importance of user perspectives in establishing content and face validity (Connell et al., 2018), practitioners were invited to reflect on their experience of using the AAG. They reported that the scale’s structure provided a unique grief profile of the bereaved participants in the study and that the items in the scale and the themes associated with them facilitated a natural entrée into the grief perspectives and troubling dimensions of a bereaved person’s grief. Two further studies were undertaken to test the AAG’s clinical usefulness, which supported the findings on content and face validity (Machin and Spall, 2006; Machin, 2007).

Phase one comprised an intersectional analysis of my practice observations and research with existing theories of grief. This resulted in a new three-category conceptual framework, capturing the range of responses prompted by loss: overwhelmed, controlled and resilient. This then led to devising a novel measure to support validation of the proposed concepts in the framework: the nine-item AAG scale. The research process, which took place in a practice setting, supported the RRL concepts. In addition, an unexpected outcome was that practitioners involved in the research highlighted the potential for the AAG scale to be used as a practice tool that could both quantitatively capture an individual’s response, and also enable a qualitative exploration of their experience.

Rosenblatt (1993) posed a theoretical dilemma: “…two contradictory perspectives recur in psychology and the social sciences. One perspective holds that humans are basically the same. The other holds that there are enormous differences among people” (p. 110). Clearly, a coherent theory of grief needs to pay full attention to human diversity while still attempting to find a conceptual frame that articulates common patterns and themes within grief (Stroebe et al., 1996; Bonanno and Kaltman, 1999). This was the challenge for me as I thought about how there might be a new approach to understanding the “range of response to loss,” while not obscuring the individual experiences and expressions of grief associated with loss. It began to evolve from an idea and an observed pattern to a more fully formed framework, which I named the Range of Response to Loss, and one which with the conceptually linked AAG could use the broad concepts to access a fuller understanding of the individual impact of loss. This essential link between theory and practice application became a significant element in the development of the RRL theoretical concepts. There was clearly more work to do in refining the RRL framework further.

It was becoming clear to me that while resilience had been identified in the RRL framework, as a positive mediating factor between the overwhelmed and controlled experiences and expressions of grief, negative or complex factors also needed to be considered as inhibitors of resilience. Vulnerability seemed to describe those factors and to be, not only, theoretically consistent with the idea of widening the concepts in a range of responses to loss, but also evident in the literature on pathological grief and the psychological risks of bereavement, and a characteristic in the presentation of grief by people seeking help.

I began, therefore, to explore theory to inform the possible refinement of the RRL framework to include vulnerability. It seemed important to begin with what had been written about psychosocial development across the life course. Erikson’s theory of human development describes the acquisition of competence as developing through the successful management of psychosocial tasks or crises (Erikson, 1980; Hendry and Kloep, 2002). A personal schema for navigating the inner and outer world of experience is described by Bowlby as an “internal working model” (Bowlby, 1969; Holmes, 1993) and by Parkes as an “assumptive world” (Parkes, 1996). Both Bowlby and Parkes were looking at these from their theoretical focus on loss and grief. Failure in the areas of personal development were seen as likely to produce vulnerability (and consequently a reduction in resilience) which will be evident in a limited capacity to deal with the challenges of loss and change. These theories provide a perspective on the etiology of vulnerability as well as resilience.

In the first development phase of the RRL resilience was identified as a primary component. The work of Seligman was used to examine the beliefs and tendencies of permanence, pervasiveness and personalization, and the descriptions associated with them that he used to described characteristic of resilience (Seligman, 1998; Seligman and Csikszentmihalyi, 2000). In his earlier work Seligman (1992) had suggested that failure in key aspects of development can produce negative perceptions about self and the world producing “learned helplessness.” This state is characterized by a pessimistic belief that bad events are always going to affect life (permanence), that setbacks in one area of life will generate helplessness in many situations (pervasiveness) and that loss of self-esteem comes from perceiving oneself as to blame when things go wrong (personalization). As with the other life course perspectives the identification of conceptual opposites, of coping or not coping with loss, provide support for those comparable concepts of resilience and vulnerability and provided some justification for their integration into the RRL framework.

The second group of theoretical perspectives on resilience and vulnerability came from the research and practice focus on personally predisposing factors and their impact on the experience of loss. I returned to Attachment Theory developments where “disorganized” attachment (Main and Solomon, 1986; Feeney and Noller, 1996) was added to the three original categories defined by Ainsworth et al. (1978). Other work on attachment led Bartholomew and Horowitz (1991) to develop a prototype of four adult attachment styles, conceptualizing a fourth category as “fearful.” The addition of “disorganized” and “fearful” to the original attachment categories seemed to me to be consistent with the proposed fourth additional component of vulnerability in the RRL model.

Further research on the Dual Process Model by Lund et al. (2010) found oscillation the least well-developed element of the model, while Shear (2010) saw loss and restoration as overlapping states rather than oscillation between the two. Oscillation may, therefore, not always represent reconciliation of loss and restoration orientation but may be a process of difficult grief resolution. Subsequently Stroebe and Schut (2016) have added the concept of “overload” to the model in recognition of the variable impact of stress and the implications this has in managing it. Refinements to the thinking and application of the model again pointed to areas where vulnerability was being added to the first theoretical proposition.

Personality related biases in grief were being explored by others such as Martin and Doka (2000). While these authors see different coping styles as related to gender, they are clear that these differences are not determined by gender. The pattern they describe has three elements: intuitive grief which is expressed emotionally and instrumental grief expressed cognitively, and a blending of these two styles to give a third category. These characteristics resonate conceptually with the RRL framework but a further element was identified. Martin and Doka saw that their category of blended grief might not always be a balanced combination of intuitive and instrumental grieving but a manifestation of dissonance where outer stoicism is undermined by inner anguish. This is a characteristic which is frequently manifest in the AAG, where a grieving person shows agreement with both the overwhelmed and controlled items, giving evidence of tension between the two; in practice providing a clear indication of vulnerability rather than blended and balanced.

Horowitz (1997) explored the impact of adverse experiences of death or loss. While the conceptual identification of intrusion and avoidance are seen as echoes of the overwhelmed and controlled grief reactions in the RRL framework, Horowitz et al. (1979) saw these characteristics for their potential to lead to vulnerability and manifest as Post Traumatic Stress Disorder They devised the Impact of Events Scale to identify the unbidden arousal of thoughts and feelings about a loss or a tendency toward excessive control of unwanted emotion, which were seen as likely to produce an unsuccessful grief outcome. – See also the work of Hodgkinson and Stewart (1998).

See Table 2 for a summary of the comparison between the RRL and other theories.

In the study of grief and bereavement there has been extensive research in defining circumstantial “risk,” as a prediction of vulnerability. Sanders (1993) had identified several factors surrounding death as indicators of risk in bereavement, e.g., sudden unexpected death, death of a child, suicide etc. Sanders also describes current life circumstances which are problematic or stressful as contributors to vulnerability. In my developing thinking on loss, I had identified the following categories of risk from the wider agendas of loss seen in social work and counseling: damaged and broken relationships; physical/mental illness/disability; unfulfilled ambitions/profound life disappointment; and social or economic disadvantage; and bereavement following unexpected or traumatic death (Machin, 2009, 2014). Others have written of disenfranchised grief whose loss is not socially recognized, e.g., grandparent’s grief, grief for an ex-spouse etc. (Doka, 1989) or non-death losses that are unresolved and/or ambiguous (Boss, 1999) and which can become chronic sorrow (Harris, 2020). The burden of traumatic death and/or stressful life circumstances are seen as stressors which may exceed the resourcefulness of an individual and adversely affect their coping capacity (Lazarus and Folkman, 1984).

Exploring theoretical developments necessary to my consideration of refinement of the RRL framework included that of avoidance, traditionally seen in Attachment Theory as an indication of vulnerability. Fraley and Bonanno (2004) challenge the absolute link between avoidance and vulnerability, and distinguish between people who are “fearfully avoidant” and those who are “dismissingly avoidant.” They suggest that the former have a difficult time in adjusting to loss, whilst the latter show a pattern of resilience. Mauss et al. (2007) make another distinction between automatic emotion regulation, where there is cognitive disengagement from emotions producing negative effects, and action-orientation where the negative effect is decreased by active problem solving (see also Koole and Jostmann, 2004; Koole and Coenen, 2007). In the Dual Process Model avoidance may be associated with restoration orientation, where individuals may need temporarily and sporadically to distance themselves from the harmful emotions and cognitions that accompany loss (Shear, 2010). This definition more closely matches the RRL concept of control as a potential indication of vulnerability, in contrast to restoration orientation being seen only as a manifestation of positive adjustments to loss and an indication of resilience. The variation in the “complex regulatory function of confrontation and avoidance” needs to recognize that positive and negative cognitive appraisals can shape both loss and restoration orientation (Stroebe and Schut, 2008, p. 5) with the implication that vulnerability and resilience constitute a spectrum of changing, rather than absolute, states. These theoretical perspectives had implications for the structure of the RRL framework and the application in practice of the AAG.

Adding vulnerability to the RRL framework (overwhelmed, controlled, resilient) meant it was important to reflect on whether the RRL should simply be defined as a four-category model or whether the structural nature of the model needed to be revised. The components of these four characteristics were beginning to emerge as reflecting two different dimensions; reactions and responses, suggesting a two-dimensional model. Several theories helped to confirm that the four components were reflecting two different dimensions of grief.

Individuals bring to an experience of loss their own varied history of experiential influences from which a personal cognitive schema is shaped (Lewin, 1935; Snyder, 1984; Epstein, 1990). Activated, as individuals adapt to new life experiences, that schema is made up of two interactive, parallel systems; one at an unconscious experiential level and the other at a conscious rational level (Epstein, 1990, 1993). Attig (2011), in his specific focus on grief, similarly distinguishes between reflexive/instinctive reactions to loss and responses which are conscious active modes of coping with loss in a process of “relearning the world.” These theories provided a rationale for the RRL to be represented as a two-dimensional model; where the overwhelmed/controlled spectrum reflects “unconscious experiential” and “reflexive reactions,” while the “conscious rational system” and the “active mode of coping” represent a vulnerability/resilient spectrum.

Dimension one: At an instinctive and spontaneous level core reactions reflect what has been absorbed in the process of personality development, socialization and life experience about the nature of emotions, thoughts and behavior and how they should/might be experienced and expressed in the face of loss. The initial grief experience, as a reaction to loss, is characterized by the crossfire of multiple chaotic thoughts and heightened feelings, sometimes accompanied by visceral physical sensations and symptoms (O’Connor, 2019), often described as the pangs of grief, along with behavior generated by this grief dynamic. The instinctive interplay of feelings, thoughts, physical sensations and behavior are represented in the first dimension of the RRL model as a spectrum of reactions from “overwhelmed” to “controlled,” i.e., a demonstration of contrasting tendencies, which give expression to the heightened experience of loss (overt) or avoid the impact of loss by suppressing grief (covert). In circumstances of traumatic loss where a person needs to function effectively, e.g., following an accident, a community disaster, providing immediate care to another person etc., the instinct for survival is heightened and control is likely to be evident.

Dimension two: As in the first dimension, the context of cultural and social learning and experience will influence the course and manifestation of vulnerability and resilience in the varied ways people navigate their management of grief and the challenges of loss. Here it is at a conscious level that people respond to the impact of loss by seeking to balance the powerful forces of emotional and cognitive distress, alongside adjustment to changes and challenges at practical, social and spiritual levels. This second dimension of the RRL is defined as a coping response on a spectrum from “vulnerable” to “resilient” (see Figure 3).

Coping in the face of loss is a journey of adjustment and reconstruction. Antonovsky (1988) identifies elements he sees as necessary for the recovery of a sense of coherence in the face of disturbing life experience: comprehensibility, where cognitive sense is made of experiences; manageability, where personal and social resources are adequate to the demands of the loss situation; meaningfulness, where cognitive and emotional processes are used to discover meaning in the loss. Martin and Doka (2000) suggest that the “reconstruction of meaning in the face of loss may become one of the most critical aspects of the grieving experience” (p. 19). A perspective reiterated and explored in the work of Neimeyer (2009) who affirms the centrality of meaning making at the heart of the grieving agenda; a task of reconciling old meanings with new ones together with finding a new sense of self. Similarly, Attig (2011) refers to grief as an active process in which we find choice in choiceless life events, and suggests that grief is a way in which we stretch into new meanings while fully honoring what was taken from us. Daunting though this may be, Joseph and Linley (2006) see the potential for “adversarial growth” in which revised assumptions may result in “existential benefit in the dark cloud of bereavement” (Neimeyer and Anderson, 2002, p. 61). These viewpoints are reflected in the RRL and its attention to resilient coping as a central focus in practice; where overwhelming emotional distress will be replaced with a new acceptance of the emotional realities of grief and the capacity to cope with it and where the aspiration for control may move from avoidance of engagement with grief to effective functioning, which accommodates the reality of loss.

The revised formulation of the RRL moved from being a framework to a two-dimensional theoretical model. It was tested using the AAG to determine whether the scale might be used to identify the new component of vulnerability and to test the AAG’s psychometric properties. Vulnerability is not separately represented in the AAG scale but it was proposed that it might be computed by adding the overwhelmed and controlled items in the scale, which are weighted toward agreement scores, with reverse ordering for the resilient items, weighted toward disagreement with the items. This would produce an Indication of Vulnerability (IV) score. Adding O + C + reversed R scores = an Indication of Vulnerability.

The research sample was made up of 168 respondents receiving bereavement support from a community bereavement service and three hospice bereavement services. Using three comparator scales, GAD -7 (Spitzer et al., 2006), PHQ -9 (Kroenke et al., 2001) and PGD (Prigerson et al., 1995) the dimensionality of the scale, its factor structure and the vulnerability factor were analyzed. The study concluded that the psychometric properties of the RRL, represented in the AAG scale, provide a measure of vulnerability based on the interactive nature of core grief reactions and coping responses (Sim et al., 2014). The results also showed that the scale affords a wider grief profile than symptom based measures alone can, e.g., the PG-13 Prigerson et al. (1995). The findings, therefore, supported the view that the revised RRL concepts could be applied in practice using the AAG scale (Sim et al., 2014; Machin et al., 2015).

However, in that research, AAG item 8– “I think it’s best to just to get on with life after a loss” was shown to be compatible with both the controlled and resilient domains. The item has now been revised and changed to—“For me, it’s best to avoid thinking about my loss.” This revision is statistically supported, now showing greater internal consistency with the concept of control, while recognizing its function may evolve from an instinctive reaction to an element of positive coping. This conceptual ambiguity reflects the theoretical recognition, discussed earlier, that avoidance in Attachment Theory and restoration orientation in the Dual Process Model is capable of demonstrating both vulnerability and resilience.

In phase two a fourth concept, vulnerability, a key presenting factor in practice, was identified. I looked at the implications of this in the light of changing theoretical perspectives on loss and grief. This then led to considering the nature of the inter-relationship between the four concepts and the consequences for the structure of the framework. The result was the reframing of the RRL into a two-dimensional model: (1) instinctive and spontaneous reactions to loss experienced and expressed on a spectrum from overwhelmed to controlled; and (2) conscious coping responses on a spectrum from vulnerable to resilient. Using the AAG, the model underwent psychometric testing, which validated the inclusion of vulnerability as a fourth concept. Furthermore, this process added evidence of the validity and usefulness of the AAG as a practice tool. This phase of the work created for me, as researcher, practitioner and developer of the RRL, a responsibility to contribute to the thinking about theoretical coherence within which diversity is not sacrificed for simplicity nor theoretical paradigms deemed too abstract for clinical application. I believe the emergence of the RRL as a two-dimensional model achieved the objectives of coherence and together with the intrinsically connected AAG scale, provide a paradigm for clinical application in research and practice (Machin, 2014).

The scale (Supplementary Appendix 1) provides four domains for practice engagement:

(a) Quantitative assessment of vulnerability, using the AAG, is based on assessing levels of vulnerability using the classification of scores derived in the validation study (Sim et al., 2014). It also provides an insight into the dynamics of grief as shown by the comparative agreement/disagreement with the overwhelmed, controlled and resilient RRL concepts. One of the underpinning therapeutic approaches appropriate for an engagement with grief is a person-centered one (Machin, 2014; Rogers, 1980) in which the voice and the individual perspective is heard and understood. The AAG provides a structure for that voice.

(b) Qualitative exploration of the story of loss and its impact; conversations prompted by responses to the AAG scale, continues the process of assessment and begins the process of therapeutic engagement. Exploring the primary themes and exploring more fully the related sub-themes implicit in the concepts, gives the help-seeking person a voice to reveal the dynamics of their grief and for telling their story of loss. A narrative approach (Angus and Hardtke, 1994; Angus et al., 1999) is central to facilitating the story of loss, engaging with it and helping it be revised in a way that can be “lived by and lived with” (McLeod, 1997, p.86). For many people, this process may provide sufficient support for them, without further therapeutic work (see Table 3).

(c) Working with the therapeutic goals, where there is heightened vulnerability, as a consequence of complex life situations and/or complex grief, focused therapeutic intervention is appropriate. Cooper and McLeod (2007, 2010, 2011) advocate a pluralistic approach to counseling, an approach well suited to working with grief. The process begins with goal setting, in collaboration with the grieving person. I have linked this element of pluralistic counseling to the important work on grief and its goals articulated by Worden (2009): acknowledging the reality of the loss; processing the pain of grief; adjusting to life without the deceased; and finding and enduring connection with the deceased while embarking on a new life. The process of adjusting to, and coping with, loss essentially involves making sense of the experience and its consequences. The perspectives of Neimeyer and Anderson (2002), Neimeyer (2009), and Neimeyer et al. (2014) on meaning-making have been centrally informative in approaches to working with grieving people. Such an approach can be blended into active interventions tailored to the individual needs of the grieving person through appropriately multiple theoretical methods, e.g., cognitive behavioral, psychodynamic, systems, meaning reconstruction etc. These theoretical and intervention approaches inform the person-centered approach of the RRL and the AAG. The RRL schema proposes that therapy addresses:

The focus is individualized by the grief needs that are evident in the AAG responses and addressed by exploration of some of the AAG themes identified in Table 3.

The AAG is a lens through which vulnerability can be addressed (steps 1, 2, and 3) before exploring those items of resilience which can be a focus for helping find direction in the building of positive coping:

(d) Measuring changes in grief over time through the re-use of the AAG at mid-points in therapy and again at the end can provide quantitative evidence of shifts in vulnerability and enhanced resilience. This is important for services as well as individual practitioners in gauging the efficacy of their work.

The AAG scale has now been adopted by many individual practitioners and increasingly as a preferred practice tool by third sector and hospice bereavement services. Use is evidenced by my engagement in consultation and training across many health and social care services in the UK over the last 20 years and correspondence world-wide with English language researchers seeking permission to use the AAG in their projects, e.g., from USA, Canada, Australia, India. Additionally, practitioners and researchers who have identified the AAG scale as a unique tool for working with grief, have sought permission to translate it into, e.g., Icelandic, Swedish, Spanish, Portuguese, Dutch, Bengali, French, Turkish, Italian, Slovenian, Ukrainian, Chinese, Hindi and Urdu.

The impetus for new practice initiatives has been driven by practitioners, who have used the AAG within palliative care settings with bereaved family members. They identified the practice effectiveness of the scale and urged its adaptation for addressing pre-bereavement loss with patients and their informal carers. The Attitude Health Change (AHC) consists of two versions, one for patients and one for carers, which were developed in response to this practice appeal. In addition to anecdotal affirmation of their value these have been tested for their face validity (Dunleavy et al., 2020; Machin et al., 2021). Further psychometric validation is needed.

Family care services also requested a children’s version of the AAG to provide a framework for assessment of need and to aid discussion with young people. A full study of the efficacy and psychometric properties of the Children’s Attitude to Grief (CAG) scale is currently taking place at Winston’s Wish (UK), a large children’s bereavement providing bereavement support and care to children and young people. A pictorial version of the CAG is also currently being produced and tested, for use with younger children and young people with limited cognitive/linguistic capacity.

Across the bereavement care sector in the UK, and promoted by the Bereavement Network Europe (BNE), there is a new impetus to define different levels of grief and appropriately varied care responses. The BNE adopts a public health model for bereavement care (Aoun et al., 2012; Macdonald, 2024), affirming bereavement as a natural life event but recognizing that this has become separated from natural support processes in Western ahead of society. This perspective seeks to address the disconnection by fostering greater public knowledge about grief. Additionally, through its identification of the varied grief characteristics which demonstrate need and risk, it sets out a tiered approach to distinguishing between the need for light-touch support through to deeper therapeutic intervention. Engaging with this perspective I am piloting a “Grief Map,” based on the RRL, as a triage tool to connect varied levels of need with appropriate intervention. Revision of an earlier RRL assessment tool (Relf et al., 2010; Brocklehurst et al., 2014) the Grief Map is derived from work on bereavement needs assessment. It incorporates three RRL based domains: core grief reactions, coping responses, and circumstantial factors (see Table 4). It is a 7-item scale with a 5-point choice of proximity between paired statements, representing resilience and vulnerability. This assessment is not based on a deficit model of grief but maps both strengths and difficulties. The Grief Map has implications for the provision of care that does not fall back on the tendency to offer counseling in all circumstances of bereavement need. It proposes a tiered plan of care and its efficacy is being tested in the pilot study.

While the predominant use of the AAG has been within practice, its attraction as a grief specific measure has attracted a wide range of researchers (see section 3.2). Many of the requests to use the AAG come from Masters and Doctoral students but the following are examples of three national studies, two in the UK and the other in Italy.

Study one looked at support needs and barriers to accessing support amongst bereaved during people the COVID-19 pandemic (Harrop et al., 2021). The same team carried out a follow-up study looking at prolonged grief during and beyond the pandemic (Harrop et al., 2023).

The statistician working with the AAG as the primary outcome measure in this research wrote, “The AAG was very useful in determining which pandemic-specific problems had the strongest impact on grief experienced by the bereaved, which was important information in planning targeted support provision. The scale was easy to use and (importantly) it has been validated and psychometrically tested. Indeed, we confirmed the high levels of reliability for the three subscales measured via Cronbach’s alpha coefficient for our data.:

The third study explored the psychometric properties of the AAG scale when translated into Italian. Results demonstrated internal consistency and construct validity. In addition, it concluded that the AAG is a valid, reliable, quick and easy to use scale that can be used both for research and clinical practice in the Italian context (Gori et al., 2023).

In 2024 a mix of 90 practitioner (n = 74) and researcher (n = 16) users of the RRL based practice measures responded to my invitation to take part in a survey about their experiences with the RRL approach to working with grief. 78% of the respondents came from the UK while 22% came from a wide range of countries from across the world. This provided a valuable snapshot of the spectrum of contexts in which the AAG is used predominantly. More than half the respondents reported that the scale has been adopted and used within their practice team, while 39% use it within their own individual practice. 80% of the respondents affirmed the AAG’s facilitation of their work with grieving people. It was clear that for many of the 12.5% who identified the need for more guidance and 12.5% who felt the scale(s) did not work as expected, there had been limited or no training in the use of the AAG or the associated scales; ensuring competency through knowledge, understanding and skill points to the ongoing importance of adequate immersion in the theory and practice processes. Overall practice efficacy and experiential usefulness can be assumed as over 40% used the AAG for between one and five years and almost 20% used it for more than five years. Qualitative comments in the survey identified the ways in which the RRL based measures, like the AAG, brought benefits to grieving service users, the therapeutic process, and to services providing care (see Table 5).

The overall impact of the RRL work captured in the responses to the 2024 survey shows that clients themselves are more able to understand the nature of their grief and show engagement and insight into how their coping capacity is changing. Consequently, key grief issues can be more quickly addressed. This in turn aids collaboration between the bereaved person and the practitioner in the process of goal setting, monitoring therapeutic change and appraising outcomes (See Supplementary Appendix 2).

Over 80% of participants in the 2024 survey expressed an interest in the proposal to form a Community of Practice (Wenger, 1998) amongst users of the RRL and associated practice scales. A venture intended to strengthen the participation, learning and evolution of best practice of the RRL and its associated practice and research tools amongst practitioners internationally. An RRL based Community of Practice has now been formed.

The need for theoretically driven guidance to support the targeting of effective support interventions in grief (Schut et al., 1997), I believe is demonstrated through the integrated connection of the RRL model and the AAG scale. This integrated connection between theory, research and practice has been pivotal in developing the RRL theoretical model.