The Lived Experience of Codependency: an Interpretative Phenomenological Analysis

Codependency is a complex and debatable concept, which has been used over the years by mental health professionals to inform their practices. Researchers have attempted to identify the main problems associated with codependency; however, their evidence is still inconclusive. This is the first time that interpretative phenomenological analysis (IPA) has been used to explore the lived experience of codependency from the perspective of self-identified codependents. Eight participants recruited from local support groups for codependency in the UK, offered in-depth information about their subjective experiences, and embedded in their lifeworld. Data was gathered through interviews and a visual method. The shared experience of codependency was portrayed by the participants as a complex but tangible multidimensional psychosocial problem in their lives. It incorporated three interlinked experiences: a lack of clear sense of self, an enduring pattern of extreme, emotional, relational, and occupational imbalance, and an attribution of current problems in terms of parental abandonment and control in childhood.

Codependency is a complex and contested concept, which has been used over the years by mental health professionals to inform their practices (Harkness 2003;Sadock and Sadock 2004;Dear et al. 2004;Denning 2010;Marks et al. 2012). It has had a strong presence in the psychological self-help literature (Schaef 1986;Mellody 1989Mellody , 1992Beattie 2011Beattie , 1992Jellen 2014).
A review of the historical development of the term demonstrated that early interpretations of codependency began to appear in the 1940s in the USA. These were associated with behaviours presented by wives of alcoholics (Price 1945;Mac Donald 1956). The development of the concept of codependency was influenced by the perspectives associated with the Alcoholic Anonymous' (AA) communities in the USA during the 1960-1970s. The influence of the AA culture in shaping the concept of codependency as an illness offered the idea that people who were close to the substance user were themselves suffering from an illness (O'Briean and Gaborit 1992). These people were viewed as enablers and coalcoholics (Cotton 1979). Codependency began to appear more prominently in the clinical and popular literature from the 1980s onward. Three models came to the forefront in this period, providing different viewpoints in codependency. These are termed and well-documented in the literature of codependency as the disease model (Whitfield 1984(Whitfield , 1987, the personality model (Cermak 1986), and the interactionist model (Wright and Wright 1991). The disease model considers codependency within the boundaries of clinical interventions and is concerned with diagnosis and treatment. The personality model of codependency highlighted the role of personality and constitutional factors in predisposing individuals to develop codependency (Cermak 1986). The interactionist model proposes a combination of both interpersonal and intrapersonal factors in the development and maintenance of codependency (Wright and Wright 1991).
Although these models form the basis of some quantitative research carried in the field (Abadi et al. 2015;Mark et al. 2012;Wells et al. 2006;Martsolf et al. 2000;Hughes-Hammer et al. 1998;O'Brien and Gaborit 1992), they arguably have a reductionist perspective, limiting the understanding of the experience of codependency within the boundaries of psychological categories, traits, and illness.
A systematic analysis of the main definitions of codependency found in the literature to date identified a thread of four elements repeatedly mentioned by the different theorists: external focusing, self-sacrifice, interpersonal conflict and control, and emotional constraint (Dear et al. 2004). However, there are no universally used definitions or diagnostic criteria, and codependency is not listed in the DSM-V (Diagnostic Statistical Manual V, American Psychiatric Association 1994). The concept has attracted much criticism due to lack of clarity, strong stereotyping, and negative labelling attributes (Gierymski and Williams 1986;Chiauzzi and Liljegren 1993;Uhle 1994;Anderson 1994;Orford 2005;Calderwood and Rajesparam 2014. The literature review demonstrated that the concept of codependency lacks a clear theoretical conceptualisation and, as a result, has generated a fair amount of discussion and contradictory evidence and theory among researchers. Most of the empirical evidence is formed by a body of quantitative research, attempting to categorise and quantify this contentious human experience. For example, across decades, researchers have attempted to identify the main psychological factors associated with codependency without clear agreement (O'Brien and Gaborit 1992;Carson and Baker 1994;Irvin 1995;Hughes-Hammer et al. 1998;Wells et al. 2006;Hoeningmann-Lion and Whithead 2007;Marks et al. 2012;Lampis et al. 2017). There has been an attempt to provide evidence for codependency in families with substance misuse problems (Prest and Storm 1988;Bhowmick et al. 2001;Sarkar et al. 2015;Bortolon et al. 2016;Askian et al. 2016). The results of these studies highlight that the evidence is still inconclusive. Codependency appears to take many forms; like a chameleon concept, it does not seem to be fully understood within the confines of pre-determined psychological traits, categories, or measurement tools.
Current studies still follow traditional routes of enquire, attempting to categorise the concept quantitatively and within the framework provided by debatable early models of codependency. For example, Chang (2018) used self-reported measures to assess the link between codependency, differentiation of self, and family dysfunction in sample of Taiwanese University students; Hawkins and Hawkins (2014) examined the relationship between codependency, assessment measures, gender traits, personality, and family alcoholism in American undergraduate students; Abadi et al. (2015) conducted a systematic review of treatment interventions for codependency and Reyome and Ward (2007) explored the relationship between childhood maltreatment and codependency in nursing students. Although offering useful insights, these studies are limited to populations of students or people with not real experience of codependency. They are also based on self-report scales, measuring traits with limited psychometric validation. Overall, they still reflect a rather limited and arguably superficial perspective of this complex experience and do not offer an understanding of the concept informed by the individual lived experience. A more in depth, qualitative perspective is needed considering the wholeness and individuality of the person, capturing the depth of their unique views, experiences, personal contexts, and narratives. This idiographic perspective is needed to complement the nomothetic perspective presented in the literature so far.
Moreover, the literature review identified a clear lack of qualitative research investigating the lived experience of codependency from the perspective of people who identify themselves as codependents. A close examination of this literature revealed that most qualitative studies were limited to a sociological perspective (Rice 1992;Irvine 2000;Blanco 2013). They were mostly concerned with social political and cultural aspects of 12 steps groups for codependency. As a whole, they offered a rather negative view of codependency, implying acceptance of unhelpful stereotyping and did not capture participants' own understandings. These studies did not address the idiographic experiences of individuals who consider themselves to be codependents and who may seek recovery groups as a way of dealing with difficulties in their lifeworlds. They left many questions unanswered and invited further psychosocial research investigating how codependency is internalised, experienced, and shared by people who identify with it. These limitations suggest that a phenomenological and idiographic study was needed, focusing on the meaning of codependency for self-identified codependents and considering their perspectives and experiences of recovery.
One of the aims of qualitative research is to offer an in-depth perspective of an individual's lived experience, which leads to a more empathic informed practice (Cassidy et al. 2011). The debates and uncertainties about the subjective meanings of codependency and the lack of research from an insider's perspective suggested that an inquiry into individuals' experiences from the perspective of the self-identified codependent was pertinent. The research study presented in this paper offered a response to this call. The aim of this research was to explore the meanings, and personal understandings associated with codependency, as understood by the individuals who identify with this concept, shown by seeking support from a recovery group, and who find it meaningful to explain the origins and development of their lived experiences. This paper furthermore will suggest how the findings of this phenomenological study might be useful to inform mental health clinical practice. This is the first time that interpretative phenomenological analysis (IPA), as a specific research methodology has been used to explore the experience of codependency. Currently, there is an increasing development of IPA research looking to obtain insiders' perspectives into mental health (Knight et al. 2003;Horn et al. 2007;Hagen and Nixon 2011) and addiction problems (Larkin and Griffiths 2002;Rodriguez and Smith 2004;Shinebourne and Smith 2008, 2010, 2011Hill and Leeming 2014). These authors have agreed that the methodology seeks both an empathic and critical understanding of the lived experience, not often captured by other forms of research.

Methodological Approach
At the initial consultative stage of this research project, it was important to determine which qualitative methodology would be most suitable to address the proposed research question: What is the lived experience of codependency among people who have sought support from a recovery group for codependents? Several methodologies were explored, for example: biography, ethnography, phenomenology, grounded theory, case study, and discourse analysis (DA). A phenomenological position was considered most suitable to guide this exploration. This is because the researchers were concerned in capturing an in-depth perspective of the individual who finds the term codependency useful to frame their lived experiences. They were interested in obtaining the insider's perspective of a number of self-identified codependents, with shared experience of attending recovery groups. For this aim, the researchers looked for methodology which addressed the detailed and specific narrative accounts of their lived experience of codependency, fostering an in-depth understanding of the complex, idiographic, and shared aspects of this.
IPA is an approach to qualitative research, concerned with the personal lived experience, and the meanings attributed by the participants, in so far as they can be interpreted by the researcher (Smith et al. 2009). It is a methodology in its own right and offers an in-depth exploration of the participants' lifeworlds. Given the negative assumptions widespread in the quantitative literature on codependency, it was deemed helpful to take an empathic, albeit questioning approach, following the argument (Smith and Osborn 2003;p.51) that IPA Bis concerned with trying to understand what it is like, from the point of view of the participants, to take their side.T here are three key fundamental theoretical principles in IPA: phenomenology, hermeneutics, and idiography (Smith 2004;Smith et al. 2010). IPA adopts a hermeneutic approach to phenomenology (Larkin et al. 2006). This interpretative component situates the IPA analysis within an interpretative circle, involving the perspectives of both the participant and the researcher (Smith 2004). Reflexivity is a fundamental aspect of IPA research (Langdridge 2007;Finlay 2008). In this study, through ongoing reflexivity, the researchers became critically aware of their position and reflected on how their personal experience, thoughts, beliefs, opinions, and interpretations, influenced the research process and outcome (Finlay 2008). For example, authors were familiar with the concept and its critiques through clinical practice and teaching in mental health.

Sampling
The IPA methodology values purposive and small samples, as there is a strong idiographic approach and commitment to in-depth data analysis (Smith 2011a, b). Specific contexts with small sample sizes are encouraged to ensure the richness of the information collected and appropriate analysis (Larkin et al. 2006;Eatough and Smith 2006). Adhering to this, eight participants (five women and three men) were recruited from local support groups for codependency in the UK. The decision to recruit from local support groups was informed by our review of the popular literature available, which informed that individuals who consider themselves to be codependents typically seek codependency anonymous groups as a way of dealing with their codependency (Beattie 2011(Beattie , 1992Mellody 1992Mellody , 1989). The support group selected for this study follows an eclectic theoretical framework underpinned by medical, spiritual, and behavioural principles.
The participants were selected based on their shared the experience of codependency. This was deemed sufficient to enhance the homogeneity of the sample, as required in IPA studies (Smith et al. 2010), although it is accepted that participants' network of other relationships would be intertwined with their lived experience of codependency. Therefore, sensitivity to participants' perspectives, relational experiences, and interpreted contexts was an important aspect of this IPA research, which aimed to capture their unique and shared experiences of codependency (Yardley 2008). This recruitment procedure was fully compatible with IPA methodology (Smith et al. 2010).
All of the participants were fluent speakers of the English Language. The inclusion criteria for the study specified that participants identified themselves as codependents. This was important because the study intended to understand subjective process, which led these participants to frame their own experiences in terms of codependency, with a shared recovery philosophy. To ensure the welfare of the participants and researcher, participants were also expected to be receiving some form of support for codependency, i.e. attending self-support groups, or receiving individual counselling or support.
See Table 1 below for more information on participants.

Ethics
The study conformed with the host university's Ethics Committee's requirements and received full approval, adhering to principles of autonomy, confidentiality, beneficence, non-maleficence, and justice. Potential participants who had their names and contact details on the group website were sent a message explaining the purpose of the study and inviting their participation. After this initial contact, if agreed, the participants received an information pack explaining the process of the study, the inclusion criteria, and the consent form. A limited amount of snowballing was also used, which included referrals from participants (Smith et al. 2010). All of the participants gave written and verbal consent. Anonymity was ensured as participants were given pseudonyms, and all identifiable information was removed from the data. The research project was informed by a team of advisors composed of three selfidentified codependents, who volunteered to offer an insider's perspective and to contribute to the research in a consultative and collaborative role, for example by advising on information sheets and interview questions.

Data Collection Procedure
In IPA studies, data collection is a dynamic process that seeks to uncover and understand in depth the participants' lived experience in so far as they can narrate this. In this study, the data collection process occurred over 6 months by means of a maximum of three in-depth semi-structured interviews and a visual method (for more information on the visual method see Bacon et al. 2017). The first author was the interviewer throughout. Each one of the interviews lasted approximately one and half hours. The repeated interviews had the intention to promote in-depth conversations over an extended period of time, as well as allowing time for the explorations of the experience through the visual methods. The visual method invited participants to bring an image or object that characterised their experience of codependency to be discussed during the second interview. This was a useful method to access a more in-depth narrative from participants and potentially encourage new insights and metaphorical understandings (Shinebourne and Smith 2011). A brief interview topic guide, with predetermined (6-10) questions was used as a guiding tool for the interviews. The schedule was designed to offer guidance regarding possible questions to be explored over the three interviews. Questions were open ended and aimed at encouraging participants to express themselves in their own words. During the participant's interview, the phrasing of some questions of the topic guide was changed or omitted, depending on the details already offered, and no particular order was followed. See Table 2 for an example of interview topic guide.

Data Analysis
The interviews were transcribed verbatim and were analysed following a distinctive, systematic but flexible process recommended by Smith et al. (2010) and Smith (2011a, b), namely: initial encounter with the text, case by case analysis, with the identification of themes, clustering of themes, refinement of clustering of themes, cross case analysis identifying superordinate themes, labelling of super-ordinate themes, and writing of a narrative report. The researchers had an attitude of openness and immersed themselves in the data, consistent with the attitude taken on data collection. The case was very much central to the analysis, and the researchers attempted to understand as much as possible about each individual case before moving to the next. At the cross-case analysis stage, the researchers remained faithful to the individual case, focusing on the lifeworld of each participant, whilst searching for commonalities or convergences of meaning (Eatough and Smith 2006). The researchers corroborated the themes inferred through discussion.
Having clarified the theoretical and practical aspects of the IPA methodology and procedure, the following section offers a detailed examination of the findings.

Results
The eight participants offered in-depth, vivid, and rich information about their subjective experiences of codependency embedded in their lifeworld. Four main themes are presented in the Diagram 1 below: Codependency feels real and tangible: BIt explains everything.F irstly, all participants revealed an understanding and lived experience of codependency as something that to them felt real and tangible, forming an important and central feature in their lifeworlds. They had all struggled to understand their enduring social and emotional difficulties until they discovered the concept of codependency. Adopting the codependency way of framing experience did not happen easily for these participants. It was gradual discovery which offered meaning to complex and confusing experiences. For them, it came as a relief to understand codependency as a socially recognised psychological problem which exerted The participants all appeared to have found in Bcodependency^a simple, singular, and allembracing explanation for a range of life difficulties and problems. For them, codependency was something so real that it felt concrete and touchable, like an illness or an underlying addiction problem, related to many forms of addictive behaviours, as illustrated by Misha: I think all addiction patterns come from codependency. Codependency is the mother ship of all addictions! None of the participants expressed stigma in relation to framing their experiences in this way. Instead, the concept of codependency offered meaning and hope that they could manage their difficulties more effectively.

The Chameleon-Self, Who Blends In
For all participants, the experience of codependency was associated with their enduring difficulties with self-concept. They all shared the experience of struggling to locate and define a clear sense of self. The participants spoke about the frustration with their lack of selfdefinition, which according to them resulted from an over-willing blending into situations. Several used the metaphor Bchameleon^to describe this process of adaptation to the social environment and relationships. Selma's quote illustrates the theme: … it is like the chameleon, you know, trying to fit in with every situation rather than allowing myself to be who I am … Participants spoke about their attempts to change and modify themselves to fit in socially, in order to feel liked, to belong and feel accepted, to gain a sense of self-esteem, yet taking this to an extreme where they lost sight of self.
Modifying myself in a chameleon like fashion to fit in, losing a sense of constancy around my values, my needs … (Selma) All participants expressed feeling locked in to subservient and passive roles within close relationships. These relational difficulties had various negative consequences; for example, participants expressed feeling overruled, staying in the relationship in spite of its detrimental and often destructive effects, and choosing partners who had problematic psychological issues. They described the experience of becoming imprisoned in their relationships and finding themselves Blocked^into these situations, feeling powerless, and unable to break free. For example, Mathias conveyed a sense of being locked into the relationship and unable to dissociate himself from his partner. He remained in the relationship in spite of feeling that it was not working. He described his over-riding sense of obligation, as something that was similar to an arduous military duty that was given to him by God.
I would be in relationships that were unhealthy, unequal, umm unpleasant umm and I would stay in them, you know, no matter what, like a marine, umm … It's my duty, God gave me this! He suggested that as a codependent, he became too adapted to each role, to a point where he would become the role and lose a sense of self. All participants described experiencing a dysfunctional degree of adapting themselves to situations as something negative and related to their codependency. Resonating with this need to subsume personal needs within relationships, most of the participants explained further various, and sometimes dysfunctional attempts to obtain a clear and better defined sense of self, as described next.
Seesawing Through Extremes in Life: BLike a Seesaw … I Feel Very Out of ControlT he experience of codependency, according to participants, was manifested through difficulties in living a balanced existence, suggesting a perceived lack of internal stability. Participants all related their lack of self-definition with continuing occupational and emotional unmanageability. Participants described the experience of lack of balance in their lives, with an excessive tendency to go to extremes of engagement in activities, and oscillating from one extreme to the other in a range of situations, for example working too much, using drugs or not looking after themselves. This was portrayed by the metaphor of the seesaw: Maybe (my life) is a seesaw, maybe is something like a seesaw, you know … I can swing from self-care to self-deprivation, self-care to self-deprivation. … And it's not very consistent, the two ends of it … if I push, and put too much weight on one end, you know, I feel very out of control, but if it is balanced, it would be easier (Misha).
Participants spoke about the experience of imbalance as a negative, portraying a sense of duality, or split; for example, stating that they felt Bup and down emotionally,^swinging Bfrom self-care to self-deprivation.^This experience of imbalance caused a sense of struggle as they searched for more stability as explained by Helena: I actually think, I needed to go down that particular path to come back to the middle, that is my experience in almost everything to be honest. I tend to flick to each end of the scale and eventually balance somewhere in the middle. They described a Bneed or urge^to do activities in what they regarded as excess and intensively, such as excessive drinking, drugs, and sex with multiple partners, seeing this as self-destructive, but associated with a need to escape feelings of inner emptiness: … I would drink too much, and then smoke too much weed, and like the sexual acting out as well … big part of the highs and the lows and all of it, just combined to it, just this craziness it was all. The majority of it was internal … a constant feeling of devastation … it's just like this paradox of devastation and emptiness … (Selma) Participants spoke about a series of possibly destructive actions and compulsions that they had eventually attributed to codependency. They described not coping well with a quiet, routine, or empty life. It is possible that they may have needed to experience the rush of activities, so as to escape this experience of inner emptiness or to experience a sense of being alive.
In order to relax I have to burn out almost, I don't know how to just relax, 'cause I somehow have to go to the extremes … Yeah, I don't work very brilliantly with the mundane; it is the steady life … … if I don't get to the edge of what it feels like to be alive, then I don't feel alive, then I get grumpy … anything that feels like life stops, it's a terrifying space … (Helena) Participants recounted what they perceived as an excessive tendency to go to extremes of engagement in activities. It appeared that, for some participants, this heightened activity was associated with a need to escape a sense of void or inner emptiness, described by Patricia as B… that feeling of a hole in the soul.T hey interpreted this as a problem related to their codependency and spoke about their struggle to establish a more balanced life experience.
Finding meaning in codependency through exploring childhood experiences: Bdown to childhood.T he last convergent theme related to participants' attributions for their problems to experiences in their family of origin. Participants had all clearly engaged in a deep analysis of their childhood experiences to provide causal attributions for their perceived difficulties, framed as codependency.
I do believe that it is down to childhood experiences and the individual child's perception of those experiences … considering that all of my siblings are messed up as well … (Selma) Participants shared a negative perception of being raised in home environments where they experienced various forms of excessive control, criticism, and perfectionism. Most recalled a rather paradoxical interpersonal family dynamic described as excessive parental rigidity and control combined with lack of support.
A closer investigation of their accounts revealed a further interesting aspect: reference to a parental figure who was perceived as physically and/or emotionally absent by most of the participants. This absence of a safe parental figure, typically the father, was portrayed by five participants and associated with their later experience of codependency.
… my father who was, quite passive, actually often quite absent, he worked, sometimes he worked in the evenings, sometimes he worked at weekends … he wasn't the men's man … my mother bossed him about, my mother ran the house … (Jonathan) The theme also conveys a sense of extreme duality, as participants described the paradoxical experience of feeling both controlled and abandoned. This duality was portrayed most explicitly by Misha, as she reflected on both her parents' contribution to her upbringing, conveying a sense of Bsplit^felt as a result of difficulties experienced when growing up: Maybe there was a sort of half factor, maybe one half of my family was not supporting me and maybe (the other half) my mother was judging me … (Misha) In summary, as participants sought to understand the difficulties they experienced in their lives they felt the need to revisit their childhood experiences. In doing this, they looked for possible faults and gaps in their upbringing as it offered meaning to their experience of the complex and varied experiences that they had come to frame as codependency.

Summary
The IPA methodology helped the researchers to understand how a small sample of selfidentified codependents made sense of experiences in their particular lifeworlds, searching for the meaning they attribute to these. The shared experience of codependency was portrayed by the participants as a real and tangible psychological problem in their lives, which appeared to follow a pattern, incorporating three interlinked subjective factors: a profound lack of clear sense of self, an enduring pattern of extreme, emotional, relational, and occupational imbalance, and an attribution of current problems in terms of parental abandonment and control in childhood. Participants reflected on the dynamic of abandonment and control as leading them to feel a sense of duality or split. They described a lack of internal instability, as they lived their lives between extremes of experience, encountering difficulties in locating and defining self, behaving like chameleons, and overly adapting to environments and relationships to obtain a sense of safety and belonging.

Discussion
The main themes which emerged from the analysis of the interviews created a rich picture of the lived experience of codependency shared by the participants and revealed participants' understandings of codependency as something real, forming an integral part of their lives. It was a concept that gave meaning to a complex set of confusing and compulsive behaviours. The interpretations suggested that the participants in part carried a rather medical understanding of codependency and found relief in encountering a diagnostic label for their previously unexplained and enduring problems. They understood codependence as a socially recognised form of addiction, which explained and offered meaning to their painful and hitherto puzzling lived experiences. This contradicted the views of early critics in the field who suggested that these individuals became labelled or stigmatised, and as such would become disempowered or lost in the sick role attributed to the label (Gierymski and Willams 1986;Gomberg 1989;Harper and Capdevilla 1990;Collins 1993;Chiauzzi and Liljegren 1993;Anderson 1994;Uhle 1994;Irvine 2000). On the contrary, the findings revealed that for these participants, codependency offered a controllable socially shared meaning for their complex and chaotic lived experiences. The concept's lack of conceptual clarity and simplistic framing device may have served to make it attractive and suitable to capture, attribute meaning, and socially validate a diverse range of experiences. Codependency served as a chameleon concept adapting to the needs of the people who identified with it.
The participants attended codependency groups, individual therapy, and read widely on the topic. They acted as Bmeaning makers^ (Langdridge 2007 p. 30), using their received understandings of codependency to make sense of their personal lived experiences. Symbolic interactionism posits that meanings are central to human life and are formed through a process of social interpretation, with people adopting the socially accepted metaphors and scripts available in their reference groups (Blumer 1986;Giugliano 2004). This is noticed by the way participants acted in relation to finding relief in the common understandings of codependency achieved in their support groups, therapy, and associated reading. Medical or psychiatric labels can be stigmatising and yet by objectifying codependency as a delineated psychosocial problem, these participants benefited from attributing a more widely accepted, socially shared meaning to their own varied and distressing life difficulties.
Social attribution theory may also be applicable as it describes people's need for a label, such as codependency to explain distressing, non-normative experiences (Heider 1958;Kelley 1973;Kelley and Michella 1980;Weiner 2008). Attribution theorists argue that when faced with adversity, people ask themselves causality questions, which in turn prompt causal searches (Kelley 1973;Kelley and Michella 1980). The theory proposes also that explanations for undesired situations are likely to be attributed to causes outside one's control to protect selfesteem (Elliott et al. 2012). Here, the participants appeared to have engaged in an ongoing process of resolving some of their intra-and interpersonal problems through their identification with the codependency label. The codependency attributions may have provided a way to explain and understand situations that happened in their lives in the past, serving also as a framework for future actions, decisions, and behaviours.
In addition, from the perspective of these participants, codependency went beyond the medical model; it was also associated with existential issues associated with problems with a sense of self. The second theme revealed the participants' struggles and search to obtain a better defined, more authentic sense of self. They portrayed some of their struggles in finding themselves behaving like chameleons, adapting and conforming over-readily to situations, which caused frustration and dysfunction in their lives. The issues associated with an undefined sense of self played an important part in the lived experience of the participants of this study; nevertheless, research in the field offered only a partial perspective on this, failing to capture the depth and significance of this experience for the self-identified codependent (Carson and Baker 1994;Crothers and Warren 1996;Irvine 2000;Dear and Roberts 2005;Chang 2018).
The IPA methodology grants that diverse and psychological and philosophical perspectives are integrated into the research to interpret and elucidate the findings (Smith et al. 2010;Smith 2011a, b). Two further traditional theoretical frameworks have been useful to interpret the participants' existential and relational difficulties identified as codependency (Winnicott 1960a(Winnicott , b, 1965aBowen 1974Bowen , 1978. Bowen (1974Bowen ( , 1978 proposed that the degree to which the person develops a cohesive and differentiated sense of self is determined by the differentiation this person obtained from the family of origin. People with low level of differentiation may have internalised a more fragile sense of their own thoughts, emotion, and needs, tending to accommodate and conform to the situations to a point that they lose their sense of individuality and authenticity. Here, we suggest that by behaving like chameleons, the participants highlighted their undifferentiated sense of self. Furthermore, Winnicott's (1965a,b) views resonate with the notion of the inauthentic, nonvalidated, and undifferentiated sense of self described by the participants. He posited that when individuals do not have their needs validated in childhood, they tend to accommodate to the needs of their parents, developing a defensive or undifferentiated organisation of self, termed Bfalse self,^similar to the chameleon type of behaviour described by the participants here. He added that people with a Bfalse sense of self^present a tendency to obtain a sense of value and esteem through excessive engagement in activities. Interestingly, this was described by the participants as they believed that their codependency was manifested in the marked occupational and emotional imbalance in their lives. They spoke about having difficulties with balance, sharing a perceived lack of internal stability, communicating a profound fragility of self, which they thought fostered experiences of intense and enduring emotional and occupational imbalance. Participants described engaging in activities to obtain a sense of esteem and validation. These participants experienced a Bsense of void,^described by one as a Bhole in the soul,^and seemed to have engaged in a frantic pursuit of activities in order to fill this. These rich accounts add further insights into some of the generalisations about codependency (e.g. emotional suppression and external focusing) previously associated with codependency in the quantitative research literature (Dear and Roberts 2005).
Finally, the study also uncovered the participants' specific social attributions of their difficulties to dynamics within their family of origin, perceived as control by one parent and abandonment by the other. Like naïve psychologists (Heider 1958), the participants appeared to have engaged also in the process of external causal attribution (Kelley1973; Kelley and Michella 1980), whereby they searched for past childhood experiences as distal causes for their identified codependency. However, the identified contradictory parenting pattern of control and abandonment was not necessarily associated with parental substance misuse as suggested by early theorists in the field (Mellody 1989;Beattie 1989Beattie , 2011Whitfield 1984Whitfield , 1987Whitfield , 1991, nor by quantitative researchers who have examined the relationship between childhood family experiences and codependency in populations of students (Crothers and Warren 1996;Cullen and Carr 1999;Fuller and Warren 2000;Reyome and Ward 2007;Knudson and Terrell 2012;Hawkins and Hawkins 2014).
The personal significance of early family interactions in the lives of participants can be better understood within Bowen's (1974Bowen's ( , 1978 and Winnicott's (1960a, b) theoretical views about differentiation in childhood. Winnicott suggested that the presence of a Bholding environment^is key to facilitating optimal psychological development (Winnicott 1960a, b, p. 591). Here, participants reflected on their family environments as negative, rigid, and unsupportive. Such environments may have prompted them to feel that they may have no other option rather than conform to an unauthentic existence and to resent the freedom to make choices and express themselves.

Critical Evaluation and Limitations
A number of quality measures were used to ensure the rigour, trustworthiness, and credibility of the study (Yardley 2008;Smith 2011a, b). The process of multiple interviewing helped to gain deeper insights and go beyond rehearsed narratives likely to have been told many times in support group meetings. In spite of this, the study is limited by factors related to methodological and participants' sensitivities. According to Shinebourne and Smith (2011), support group forms a good sample representation, as it adequately meets the IPA criteria of a purposive, context-specific, expert knowledge group. This recruitment procedure was planned with the positive intention of increasing homogeneity within the small sample; nonetheless, it may have limited the study in some ways. Limitations around received, rehearsed, and edited narratives had to be considered, as this group appeared to be well informed about lay and psychological theories about codependency, for example they attended groups and read widely about codependency. It was not possible to entirely disentangle Bfirst hand^experience from the framing applied from this learning within the support group and beyond. Despite its aim to stay experience-close, IPA research into other contested conditions, such as chronic fatigue syndrome and addictions, has faced similar issues (Dickson et al. 2007(Dickson et al. , 2008Shinebourne and Smith 2008). However, concurring with other research carried out in support groups (Shinebourne and Smith 2011), participants' accounts varied considerably, and were not in any obvious sense scripted.
In addition, although the sample was large for IPA research (Smith et al. 2010), it could nonetheless be considered relatively small and specific, and therefore findings are not straightforwardly generalizable to other contexts. The participants volunteered to take part in the study, so they may have been highly motivated to share their experiences and understandings of codependency. Other self-identified codependents may have different views and experiences, especially those who have not yet encountered or found value in support groups. This is a qualitative study, so we are not claiming that the findings can be generalised simplistically, in particular to people who identify with codependency yet do not attend formal support groups. We nonetheless argue for a theoretical transferability (Smith et al. 2010). Smith (2011a, b;p7) suggested that the particular aspects explored through IPA Btakes us closer to the universal.^Therefore, we propose that the findings may guide further research and inform clinical practice.

Conclusion and Clinical Implications
It is important that therapists working in this field need to be sensitive to the importance of this label, and the attributions that it implies, to this client group. Receiving such a label may not necessarily come with negative connotations despite some previous sociological and feminist debate on this matter; on the contrary, it may offer a socially recognised explanation for complex, enduring, and distressing life experiences. The findings highlighted the need for therapists to be aware of issues related to fragile self-concept and the experience of problematic parenting experiences in childhood when offering support to codependents. People with codependency issues would benefit much from empathic listening as it can bring a sense of acknowledgement and validation of their life experiences, contributing to the restoring of their sense of self. A thoughtful therapist-client relationship, focusing on embracement, holding, and support thereby proving a stable secure therapeutic environment could help these individuals as they engage in a process of self-construction. Therapists working with these clients should be conscious of their extreme oscillation in engagement in activities. They can plan psycho-educational interventions, which are aimed to assist these clients to consider a more balanced lifestyle. Stress management techniques and emotional regulation strategies can be offered to assist to regulate their emotional and occupational instability. The extremes of oscillation in feelings and activities may also create barriers to forming a strong therapeutic alliance and therapists may need to use supervision to manage the emotional effects of such oscillations on self.
It is hoped that the results of this study will provide a base for developing a more empathic and contextualised understanding of the experience of codependency, which in turn will enable mental health professionals to offer support which is relevant to these individuals' experiences.

Compliance with Ethical Standards
Conflict of Interest Ingrid Bacon, Elizabeth McKay, and Frances Reynolds declare that they have no conflict of interest. The study did not receive external research funding.
Informed Consent All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). Informed consent was obtained from all patients for being included in the study.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.