10 A one-year follow-up study was conducted on a cohort of adults with a moderate to severe ABI admitted to the neurorehabilitation centre in the northern region of Denmark.1 The inclusion criteria were: age 18- 66 years; residence in Aalborg, Jammerbugt, Vest Himmerland, or Bronderslev municipality; and admission to Bronderslev Neurorehabilitation Centre (BNC) in 2013-2014. Forty-five adults met the inclusion criteria. However, one died between the first and second interview and one decided not to be part of the study after all. This resulted in a total pool of 43 participants.

11 All participants were interviewed twice. The first interview took place in the acute phase during hospitalization (in 2013-2014), and the second interview took place one year later (one year post-injury) in the informants’ homes (in 2014-2015). The first interview was a short conversation during the pretesting at the hospital and was reported as field notes. The short conversation form was chosen at this time out of ethical concern for each participant’s physical and psychological state. The second interview was approximately 45-60 minutes, and was tape- recorded and later transcribed. Although the length of the two interviews was different, the same interview guide was used, although shortened in the first interview. Both interviews were semi-structured in order to allow some level of improvisation: for instance, small talk or comments about the surroundings or communication with relatives in the participant’s home.

12 The first author conducted the first interview and both authors were present during the second interview, one observing and the other actively interviewing. Although this study is qualitative, the authors also refer to test results on depression, measured by the Major Depression Inventory (see Bech, Rasmussen, Olsen, Noerholm, & Abildgaard, 2001) and quality of life, measured by the World Health Organization (1996) Quality of Life-Bref questionnaire. However, in this case study, these quantitative results are interpreted qualitatively.

13 The narrative analysis involved two phases: a descriptive phase and an interpretive phase. We started with a thorough reading of the narrative accounts, examining language, sentences, structure, function, and content in order to identify and highlight key features. First, the interview transcripts were carefully read and initially analyzed separately by the two authors. This was followed by an interpretive phase, during which we produced a joint analysis by using Bamberg and Georgakopoulou’s (2008) analytical approach. This two-step analysis was done to improve validity. We encouraged participants to tell us about their experiences. Therefore, there was no predetermined framework of meaning, because meaning was constructedthrough storytelling. An important limitation in adopting this analytic strategy is the selectiveness that is required when choosing which narratives to consider as important. Nevertheless, the focus on the rich production of stories encourages the researcher to be open to any form of meaning construction. In this article, a case study is presented involving a 27-year-old Danish woman. This case study allows us to go into depth regarding one of the key patterns reported by almost all the participants: the psychological changes and identity transition that emerge over time.

14 In general, case studies have been criticized for their lack of representativeness. We acknowledge that the research produced in this case study is both situated and context-specific, which means that it cannot claim universality. However, we point to the notion that the challenges of identity during different transitions in the recovery process were not only relevant to this particular case study, but for the total pool of 43 participants. Therefore, this single case study is representative of the psychological changes and identity transitions that emerge over time, as found in this study as a whole.

17 The first author meets Mette for the first time during spring 2013. Mette has suffered a stroke just three weeks earlier and is now being observed at a rehabilitation centre, following the acute hospitalization phase. Mette is pregnant and is to give birth in a month to her first child. During this first meeting, Mette tells a story about being happy and relieved because she got off so lightly (with the ABI). She often finds it hard to look at the other clients at the rehabilitation centre who are more severely affected by ABI. She seems happy and cheerful, and smiles during the interview. Mette is being discharged later that day after only two weeks at the rehabilitation centre, since she has no physical consequences. She agrees with the centre’s evaluation; she does not feel any severe consequences from the ABI. During the testing (for depression and quality of life) she mentions that a rehabilitation plan has been made, but Mette says, “There is very little help I need.”

18 The tests disclose a high quality of life, with no sign of depression. A nurse from the psychiatric ward (pediatric) will get in touch with Mette when she is about to give birth to her son. The doctor has arranged this to prevent any delayed psychological reactions during or following labour. However, Mette does not report any psychological consequences so far. On the contrary, she seems relieved, and during the testing she says, “Well, you won’t find much here.”In connection with the rehabilitation, she explains that she has never felt “lost” (in the sense of getting what she needed), and that she has received all the support she could wish for.

19 In this excerpt we see a young woman, soon to be a mother, who positions herself as lucky that she has experienced hardly any consequences from the ABI. We see how Mette constructs herself as different from the other clients with an ABI (identity dilemma 2: sameness and difference) and how she got off lightly compared to them. In this way, Mette constructs herself as the same (constancy) before and after the ABI. She positions herself as the one who got off lightly and, therefore, hardly feels any change (identity dilemma 1: navigation between constancy and change). The first author also co-constructs Mette as happy and even cheerful in the accounts describing her.

20 The second meeting takes place about one year later. A lot has happened in Mette’s life since the first meeting. She has become a mother and moved into a new house with her husband, Christian. Mette is still waiting for clarification in regard to her return to work. The interview2 begins with a question about how Mette experiences the ABI today:

In this first excerpt, Mette is focusing on the physical consequences. Later in the interview, Mette addresses the psychological aspects as well and says, “It is really, really big psychologically.” Becoming a mother and having an ABI was especially very difficult:

21 When Mette is asked what it was like to become a mother, she answers in terms of two specific terms: “anxiety” and “bad conscience,” of which “bad conscience” is repeated three times. By doing so, Mette constructs her position as a mother in a negative way. She continually uses the words “I could not,” thereby positioning herself as a passive and incapable mother and in opposition to her son, who is positioned as perfect and innocent. Meanwhile, it is interesting that Mette, at the beginning of the interview, points out the physical complications and not until later in the interview says that “I am not that affected physically. It is really, really big psychologically.” The depression test conducted at the second meeting indicates that Mette shows signs of severe depression. Furthermore, there is a significant decrease in her psychological and physical quality of life compared to the pre-testing one year earlier.

22 From these excerpts, we see a young woman who, in comparison to the first meeting, now suffers from the ABI in a highly psychological way. Mette constructs herself as an inadequate mother, both in relation to her son (not being able to nurture him) and according to social norms (not being part of the maternity group). Thereby, Mette constructs herself as different from others (identity dilemma 2: sameness and difference), but at this second time, the difference from the first position is a negative one. Furthermore, she constructs the transition from being social to living in a box (identity dilemma 1: navigation between constancy and change) as hard. This construction points to the conclusion that Mette cannot cope with the changes and the navigation of these identity dilemmas. For this reason, we need to take a closer look at the rehabilitation process, in order to understand what might have negatively influenced the development of Mette’s sense of self.

23 In the previous section, we saw how Mette explains the physical complications as the answer to how she feels about having a brain injury. Meanwhile, the following description shows that the psychological consequences affect Mette's life as well:

24 In this excerpt, Mette positions other people as blind to the ABI: “people could not see it.”Furthermore, she continues and nowincludes herself in this position: “And I could not really see it.”To take on the position of not being able to see “it” could relate to the question of acceptance and how to accept something which is invisible. The hidden and invisible consequences of an ABI are then used as an argument for the conclusion she draws: “Then they [professionals] could not really help me.” Drawing on Bamberg’s theory of discourse perspectives (Bamberg, De Fina, & Schiffrin, 2011), the dominant discursive practices centre on the formation of a consensus of what is agreed upon and taken to be truth. When Mette refers to the fact that people, including herself, could not see it (the ABI), she draws on a master narrative saying that brain injury rehabilitation is mainly related to physical and visible consequences. In this way, Mette’s accounts of the physical consequences are a way of legitimizing the claim that something is wrong. Based on this analysis, we find that Mette constructs her ABI primarily at a physical level.

25 Next, Mette talks about her previous job, where she worked as a healthcare service provider:

26 In order to understand what happens here, we draw on the first two steps of positioning analysis: how characters are positioned within the story, and how the speaker/narrator positions herself (and is positioned) within the interactive situation (Bamberg & Georgakopoulou, 2008, p. 8). When Mette talks about her previous job, she makes an interesting shift from talking about her position (“what does a person do”)as an “assistant” for elderly people, to suddenly including herself in this position as a patient (“what do we do?”; emphasis added). In this excerpt, we see how Mette shifts from positioning herself as a service provider, to suddenly being the one provided for. Still, it is interesting to see how she makes a kind of knowledge transfer when she uses her previous experiences as a professional in order to arrive at the conclusion about compensating in order to experience quality of life. Following Bamberg, the notion of small-d discourse can be used in order to interpret this quotation as a sign of wanting to be an agent, someone who takes initiative and actively constructs discourses and positions rather than being a passive recipient of those constructions. It shows a way of trying to progress and not regress, since Mette can use knowledge from a previous position to inform a current position.

27 Therefore, the authors argue that this way of using knowledge contributes to and generates an identity process in which Mette becomes aware of her professional identity and can use it constructively in her recovery process. Meanwhile, the small-d discourse (agency) should be viewed as situated and not constant or universal. For instance, in the next excerpt, we see a shift from small-d discourse to capital-D discourse. This happens when Mette is asked about her future expectations:

28 Mette uses the phrases, “what is going to happen with my life,” and “what will happen to me, and what will happen to us.”In using these words, she accepts and actually takes on a more passive position of waiting and being in a standby position.

29 Level two in the positioning analysis concerns how the narrator positions a sense of self/identity with regard to dominant discourses or master narratives. In the beginning, we saw how the dominant medical approach to rehabilitation constructed a physical focus. Moreover, we saw how Mette took on this position, and thereby positioned herself as helpless. Mette is positioned by the system as a passive recipient when she is waiting for a future clarification. However, Mette still needs to either accept or agentively deny this offered position: that is, she has to actively choose how she wants to reconstruct her narratives and identity in relation to this offered position.

30 So far, we have seen how Mette constructs her ABI, and how discourses play an important role in this process. Now, we take a closer look at how all of these aspects are brought into action and constructed as narratives. Next in the interview, Mette is telling us about the period of time when she met her husband, Christian. This was a time where she engaged extensively in social activities with her friends. Mette constructs herself in this period of time as follows: “I was the free Mette” (emphasis added) and continues:

According to the three identity dilemmas, Mette is here struggling with the dilemma of constancy and change. For instance, she says that she is the social Mette (constancy), in the temporal form of presence, then she elaborates on how she cannot go out and have drinks with her friends anymore and how they miss the “old Mette.” Here, we identify a lost self: the social self. We find that the reconstruction process is characterized by much ambivalence. For instance, Mette says: “You see, it is a new Mette, that has come […]. Who is this new Mette, who is about to come, and will my friends accept the new identity, and will they still see me as who I was before?”

31 Ambivalent accounts and different scenarios are presented here. Mette worries about whether her friends see her as she was “before” the ABI. At the same time, she wonders whether her friends will accept the “new”identity. Furthermore, she is actually presenting three different identities: a Mette that “has” come, a Mette who is “about” to come, and finally, the Mette she “was” before the ABI. These accounts show that Mette is navigating between different identities: who she was, who she is, and who she will become.

32 When she asks, “Can my friends accept the new identity, and will they still see me as who I was before?” Mette also constructs a potential risk of losing her social and personal relations. Earlier, she constructed the loss of the old Mette, in saying that both she and her friends “miss the old Mette.” Moreover, she says: “He [her husband] fell in love with a Mette who is not there anymore. And how does that affect his love for me?” Mette preferably wants to position herself in a way that would help her maintain her social and personal relations. The analysis shows that this is also why Mette cannot quite accept this developing identity (the “new Mette”), and therefore, she holds onto who she was (the “old Mette”). However, by constructing such a hope of going back, she also constructs a possibility of being the same and, thereby, being part of the same social environment that she once belonged to. In this case study we find that the construction of a lost self has to do with being the same, but different.Mette reckons that she is not the same; however, she does not know who she will be in the future.

33 In sum, the analysis points to different factors influencing the loss of identity and the identity reconstruction process. First, we saw how master narratives in rehabilitation practices enter into the process of identity reconstruction. Later, we found that Mette on one hand wishes to construct herself as she was before the ABI, but still has to reconstruct herself and accept this position as different. This is necessary in order to integrate the three presented identities (a Mette that was, that is, and that is to come) into a more coherent narrative.

34 In the analysis, we point out a developmental process both in the navigation of identity dilemmas and in ways of (re)constructing identity (in forms of narratives) just after ABI and one year post-injury. We will now discuss the challenge and complexity of the identity dilemmas. In addition, we will link the challenges of identity reconstruction to a continuing lack of psychological focus in contemporary rehabilitation practices.

35 In the first part of the analysis, we saw how Mette constructed herself as the same just after the ABI and different from others with ABIs. By doing so, she was somehow aware of the dilemmas at stake (“have I changed?” and “am I different from the others with ABIs?”). Later, we saw how Mette one year post-injury is in the middle of an identity reconstruction process, where guilt, loss of identity, and ambivalence are recurrent themes in her everyday life. Mette struggles with an uncertainty regarding who she is, how others see her, and who she will become. Just after the ABI, her need of professional help is constructed as very slight, which can be caused by two things: Mette has already received a great deal of help, and therefore her needs have changed and reduced; or the “biggest challenges” have not occurred yet. These two possibilities can be considered in combination. However, given the analysis, we find the latter explanation most likely. Furthermore, this is consistent with the literature within the field, which shows that the psychological consequences following an ABI often should be considered from a long- term perspective (e.g. Astrom, Adolfsson, & Asplund, 1993; Lezak & O’Brien, 1988; Prigatano & Summers, 1997).

36 According to Bamberg, De Fina, and Schiffrin (2011), identity is a personal and social matter and, therefore, social contexts and interactions are highly central in the reconstruction process following an ABI. In this case study, we saw how Mette’s friends also co-constructed a loss of the “old Mette,” saying that they “miss the old Mette,” a position Mette also adopted. We thereby see how the social environment actively participates in co-constructing the identity dilemmas (identity dilemma 1: telling Mette that she is not the same anymore). The identity dilemmas all interact in a reciprocal way. For instance, when Mette constructs a potential fear of changing, since this might mean losing her friends, a new identity dilemma about being the same or different from others comes into play. This new identity dilemma about sameness and difference is important to Mette in terms of being socially included or possibly excluded in the future. It shows that the identity dilemmas are part of a constant mutually and socially constructed process.

37 Still, according to Bamberg (2004), identity also involves personal dimensions. This can be related to how the individual manages the development of the identity dilemmas: for instance, how Mette copes with and positions herself with respect to the reactions and statements from family, friends, and the maternity group. Therefore, the authors argue that agency is an important factor to consider in the management of these processes. Agency refers to actions based on free choice and initiative, which can be seen in language, intentional expressions, and actions (Bruner, 1991). From a discursive point of view, it can be understood as the individual’s ability to be either co-constructor or de- constructor of discourses (Bamberg & Georgakopoulou, 2008). The third identity dilemma, the management of agency between Bamberg’s (2011) “double-arrow of a person-to-world versus a world-to-person direction of fit” (p. 3), has a potentially important bearing upon how the other two identity dilemmas are managed. To be an agent is to have the ability, will, and initiative to construct one’s own discourses and not just accept the positions and discourses offered by society. Furthermore, agency can be understood as a way of trusting oneself and as the very important distinction between constructing one’s life as something that happens to one (passive construction) or life as something one has influence over, formed by one’s own actions and intentions (active construction). Whether or not one has a high level of agency becomes interesting when facing challenges and obstacles in life: for example, how the navigation between constancy and change and the establishment of a connection between sameness and difference is managed.

38 In practice, the importance of agency would mean that professionals cannot, and should not, solve the identity dilemmas. Instead, the focus should be on strengthening the level of agency and re- establishing hope in individuals with an ABI, and thus supporting and empowering them in their identity (re)construction process. We argue that this will lead to a more successful way of coping with the ambivalence and complexity of the identity dilemmas.

39 As already stated, the psychological challenges facing rehabilitation often develop after some time. For this reason, professionals should not rely on momentary images of people with ABI and assessments obtained during the acute phase. For instance, a psychometric test of depression or quality of life is in itself not a sufficient predictor of future depression or change in quality of life for the client. Therefore, we argue that the psychological consequences associated with rehabilitation should be considered in a broader sense, both in relation to time and to the interventions used. Just as narratives develop over time, so does the need for psychological support.

40 The ability to construct a coherent story about a person’s illness and rehabilitation process is said to be a critical component in the healing and recovery process. The anthropologist and narrativist Mattingly (2005) describes how the storymaking or narrative is an influential factor in reconstructing identity: “To tell a story about your illness can have a healing effect” (p. 15).

41 On the basis of the analysis presented here, the authors support the view that narratives can be considered as a way of coping with and comprehending difficulties in life. Therefore, a narrative approach can have a positive or facilitative function. Meanwhile, this case study gives us reason to reflect on how narratives also can become retraumatising. By this, we mean that stories, for instance, by the use of language and metaphors, can lead to negative self-conclusions that are not challenged. Retraumatising narratives can be considered on two levels. First, they can be related to the level of agency, with reference to how individuals intentionally manage the identity transition, distancing themselves from self-stigmatising conclusions. Here, we saw a development in Mette, who said, “There is very little help I need,” and one year later, “What will happen to me?” Secondly, retraumatising narratives should also be considered in relation to the already mentioned lack of psychological rehabilitation. This case study points to the conclusion that changes in narrative style over time may correspond to an increased level of distress. This was found not only in the single case study, but also in the majority of the 43 participants. Therefore, not receiving any support in the identity transition process indirectly becomes a co-constructive element with respect to how the individual constructs the ABI and positions herself, as illustrated in the case of Mette. In other words, reconstruction of identity takes more than the ability to be an agentive self-narrator. With reference to Bruner (1991), narratives are not designed only for happiness or positivity, but should also be allowed to contain ambivalent and complex material. Still, individuals develop a sense of change, challenge, and, as we saw with Mette, a loss of self after they have lived their post-morbid lives for a longer period of time. This sense of change, challenge, and loss of self can lead to an amount of psychological distress, which then in turn leads to a reduced flexibility and sense of agency in relation to the identity transition process.

42 This case study resonates with some of the findings in the existing literature on ABI and identity: for example, Nochi’s (1997, 1998) findings about a lost self. However, the “void” or the loss of memory that Nochi refers to in his study is not found in this case study. On the contrary, Mette’s narration and memory of who she was seems to challenge a successful navigation between sameness and difference, and generate a feeling of a lost self. This case study illustrates that even without the “void,” there can still be the construction of a lost identity. Therefore, we find that there are different mechanisms involved in how memory or loss of memory challenges a person’s self-understanding.

43 Fraas and Calvert (2009) investigated factors leading to successful recovery and productive lifestyles after an ABI. They interviewed 31 ABI survivors. Four subthemes were found relevant to the ABI survivors: development of social support networks, grief and coping, acceptance of injury and redefinition of self, and empowerment. These themes can be identified in our case study as well. Mette’s identity reconstruction is closely related to the identity dilemmas: who she was, who she is, and who she will become. In relation to social support networks, a pattern can also be recognized. Fraas and Calvert found that participants were not always able to maintain social support networks during the initial recovery period, especially with friends. Seventy-one per cent of their participants felt that social support networks began to deteriorate during their recovery process. In this case study, Mette also said that she missed going out with her friends, and that she could not maintain contact with all her acquaintances as she used to. By the terms grief and coping, Fraas and Calvert refer to participants reporting a progression though several stages of grieving before they were able to accept their post-injury self. In this case study, the participant metaphorically illustrated how: “The tiredness and the ABI … I have to be best friends with it, and walk hand in hand with it for the rest of my life. But damn, it’s hard to be friends with!”Finally, Fraas and Calvert found that empowerment played a role for all the participants, contributing to a productive life. Empowerment is described as an ability, that is, to reintegrate into communities and to set one’s own goals for achievement. However, the informants in Fraas and Calvert’s study reported that, in the early stages of their recovery, decisions were often made by family members, schedules were made based on therapy routines, and choices were often based on functionality. This concept of empowerment is very comparable to this case study’s concept of agency. This finding confirms that agency is very important and central for people with ABI in relation to living according to their own initiatives, reintegrating or rehabilitating, and also rejecting or challenging capital-D discourses. These findings together with this case study, suggest that people with ABI have to navigate difficult personal and social changes.