Participants

The Clinique des Maladies Mentales et de l’Encéphale (CMME) at Sainte-Anne Hospital, Paris, France, is an adult psychiatric department providing specialised care for Eating Disorders (ED) and comprising an outpatient and an in-patient ED unit. Patients may be referred to this department by their psychiatrist, their general practitioner or even themselves. They may come from metropolitan or overseas France.

Our study sample was a consecutive series of: (a) female patients, (b) aged at least 18, (c) with anorexia nervosa as defined by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition [14] and (d) admitted to the ED in-patient unit between January 2008 and December 2010.

Our sample consisted of 63 women with anorexia nervosa, 40 with anorexia nervosa restrictive type and 23 with anorexia nervosa binge-eating/purging type. Mean body mass index at admission was 14.7 kg/m² (SD = 7.0). Their mean age was 27.6 years (SD = 9.0). Fifty women (82%) were unmarried, and only 5 had children. In all, 27 (42.9%) were students, 16 (25.4%) had jobs, and the others were looking for work or on sick leave.

Ethics Statement

For the present study, all the data used have been collected in the normal course of care and as a part of the regular admission procedures. Patients were informed that data were to be used in future studies and verbal consent was obtained, which was enough to fulfil requirements from the French Law on Bioethics. Also, data computerization was approved by the French commission on data computerization and personal freedoms (Comité National Informatique et Libertés, CNIL) and The French Ethics Committee for medical research (Comité Consultatif sur le Traitement de l’Information en matière de Recherche dans le domaine de la Santé, CCTIRS).

Measures

Data analysis

Application letters were extracted from medical files and provided, without any names attached, to the research team for qualitative analysis. Sample size was defined by criteria of data saturation.

NVivo 9 qualitative data analysis software [16] was used to assist our data management and analysis. All letters were analysed by interpretative phenomenological analysis (IPA), with word-processing and spread sheet software. Using this approach, no pre-existing theme is imposed on the data but themes begin to emerge as the analysis takes place. The data analysis procedure was thus inductive. The literature was reviewed after this step. First, two authors (PG and ARL) specialized in qualitative research, repeatedly read the letters (in their original language, i.e. French). Concept codes were developed and initial themes were identified and annotated in margins. This was close to being a free textual analysis. Each reading of the letters had the potential to bring up new insights. Second, recurrent themes, defined as themes reflecting a shared understanding of the phenomena in question among participants [17], were identified across letters. This second step involved a more analytical ordering, as researchers tried to provide a meaning to the connections between themes. Some of the themes tended to cluster. The aim of this dynamic and cyclic process was to recognize ways in which participants’ narratives were similar but also different. Thus researchers integrated new emerging themes, to take convergences and divergences in the data into account [18].

To enhance validity, the analysis was carried out by two authors (PG and CN). Moreover, the letters were read by one other author (ARL) to improve consistency and coherence of the IPA. The research team met repeatedly, to ensure that the themes identified were an accurate reflection of the data and that the analysis was not confined to a single perspective. The meetings helped to highlight necessary clarifications and modifications among the themes identified [18].

Extracts of letters have been selected and are provided below, in order to exemplify each underlying theme. The letters have been freely translated into English for the sole purpose of this article. This translation tried to preserve the essential meaning and content of the letters and, as far as possible, their general tone.

Loss of Control

The participants described the transition from the impression of control their disease initially conferred to the loss of control that followed. The participants explain that at the beginning of the disease, this feeling of control made them feel valuable and that progressively, they came to realise that it was illusory. In the letters, the expressions ‘vicious circles’ and ‘infernal spirals’ recurred frequently.

Intensity of distress

The women described the intensity of their suffering and mental exhaustion. Another aspect of their distress was the extreme loneliness.

Ambivalence

Awareness of ambivalence seemed to encompass two aspects: inner struggle and fear of recovery. These two themes, although distinct, may also be interconnected. That is, inner struggle may reinforce the fear of recovery and vice versa.

Main findings

In this qualitative study, 63 patient application letters have been used to identify the principal subjective themes associated with a request for in-patient care in a specialised treatment centre for anorexia nervosa.

Six principal themes have been identified; they can be regrouped into three domains of experience: perception of the loss of control of the symptoms, intensity of distress and awareness of ambivalence. These three domains of experience that together drive the women towards asking for help and more precisely for in-patient treatment reveal a process of awareness at three levels. First, the patients seem to be developing an awareness of their loss of control over their symptoms. This result is interesting because it echoes the need for control that characterises individuals with anorexia nervosa. The feeling of control that they perceived at the beginning of and throughout the disorder has progressively been replaced by a feeling of helplessness and impotence in the face of the disease. At this stage, the women appear to be able to ask for help, especially to regain this capacity for control. This result is in line with what Reid et al [19] have evidenced in a qualitative study of anorexia nervosa patients seeking for outpatient treatment: ‘the struggle to retain control resulted in some feeling desperate and suicidal, leading them to seek treatment’. The second level of awareness is this of the suffering and its intensity. Here, again, awareness of the distress linked to the disease echoes the initial phase of well-being engendered by weight loss. More generally, awareness of the suffering is the counterpoint to the initial positive self-reinforcement process, as Serpell et al showed [20], describing the major disease-linked benefits anorectic women initially feel, such as ‘feeling protected, gaining a sense of confidence, and feeling different’. Similarly, Nordbo et al [21] hypothesize that anorexia nervosa may procure a ‘feeling of stability, security and mental strength’ for women with this disease, thus enabling them to ‘avoid negative emotions’. Finally, the third dimension corresponds to the awareness of their ambivalence and specifically of their inner struggle. It appears that the patients realise they cannot cure themselves all alone, that they actually need specialised help and that this ambivalence makes outpatient care difficult. This awareness of the need for help again echoes the propensity of patients with anorexia to associate their desire for autonomy with doing ‘it all by themselves’. Indeed, anorexia nervosa often begins at a moment in their lives when young women begin to want to become autonomous. At the beginning of the disease, anorexia nervosa allows them to affirm themselves, to differentiate themselves from others and to create a distance from their family that gives them a feeling of independence [22]. Anorexia nervosa also has a role in providing an identity – the “anorexic identity” – as reported by Jenkins et al [23]. Patients who have recovered explain that the disease defined them as a person and that the recovery process implies the management and the change of this ‘anorexic identity’.

The three levels of this awareness thus conflict point by point with the initial experience of the disease and especially the feeling of control, the absence of distress or complaints, and the demand for autonomy and independence. Our study suggests that patients request in-patient treatment when the initial positive reinforcement and maintenance processes are no longer effective.

Another interesting result is that the reasons mentioned by the women for the request for in-patient treatment were eminently subjective and personal. Their request for admission to hospital was part of a personal process that seems to follow their personal experience of their increasing awareness. This result is the opposite of what has been suggested by earlier studies, which indicated that women agree to be admitted because they give in to pressure from their families or caregivers [24].

In addition, it could be noted a discrepancy between the medical priorities and the patients’ perspective. For the women, the decision to be treated as an in-patient appears to be focused on an internal process by which they become aware of their loss of control, of the intensity of their distress and of their ambivalence. On the other hand, the medical criteria for hospitalisation focus more on the physical emergency [8]. This discrepancy in physician and patient priorities has already been described by clinicians [25,26], but never previously demonstrated by the discourse of the women themselves. According to Guarda et al [25], individual motivation for treatment is linked to their desire to obtain a ‘moment of respite from the physical and psychological consequences of their disorders but without gaining significant weight’. The doctor-patient encounter is successful when they share the same priorities.

Finally, the absence of a theme related to the body itself is surprising and noteworthy. The body is approached only with medical terms, very factually, and solely to describe treatment. No emotion, no sensation is attached to the body. This distancing of the experience of the body and the emotions has been described in the literature. In particular, some authors suggest that disorders of self-perception may reflect a process of displacement to the body of unexpressed emotions or threatening feeling [27,28]. Moreover, this perceptual distortion is not frozen in time, but fluctuates as a function of the women’s emotional status and may be enhanced during difficult emotional situations [29]. If the perceptive distortions of the body enable negative emotions to escape, one can suppose that this symptom maintains the disorder and is therefore not a reason for a request for care.

Clinical implications

Our results may have several clinical implications.

First, our results emphasize the importance of motivational work with patients. Although motivational interviews were initially developed in the fields of substance abuse [31], they have a key role in the management of anorexia nervosa [32]. Their objective is to diminish ambivalence and resistance to change and to increase commitment to treatment [33]. During motivational interviews, clinicians use a ‘decisional balance’ tool, where the perception of the disadvantages in the current situation can engender a desire to change. Regarding anorexia nervosa, several studies have suggested that motivation for change is predictive of a better short-term prognosis [34,35]. Similarly, Espindola et al [36] report that ‘motivation and stimuli to remission’ are a major theme that emerges when patients in remission from anorexia nervosa are interviewed about it. The results of our study might allow clinicians to guide their interviews toward themes that are important for the patients, namely, loss of control, distress and ambivalence. Targeting these themes during interviews may help patients come to terms with their situation, at the moment when in-patient care becomes necessary. Moreover, the arguments advanced by physicians for hospitalisation could focus more on the patients’ subjective experience than on medical and somatic arguments.

A second valuable and practical aspect of knowing the patients’ own reasons for seeking in-patient care is to reduce the number of premature departures (dropouts) from specialised ED units. Premature departures account for 20 to 58% of admissions to these units [37] and have an increased risk of relapse at one year and of long-term chronicity [38]. One can suppose that having these individuals face their own motives for asking for admission signifies to them that they play an important role in their hospitalisation, which could have a positive effect on the continuation of care. This corresponds to the concept of ‘human agency’ [39], that is, people’s capacity to recognise themselves as the author of their acts, to attribute their ideas, actions, and feelings to themselves. Two previous studies have shown a low level of agency in patients with anorexia nervosa [40,41]. In-patient treatment could thus focus on helping patients perceive themselves as the agent of their treatment plan, with their own reasons for seeking this care.

In conclusion, requests for in-patient admission by patients with anorexia nervosa were found to be motivated by very personal, subjective experiences, unrelated to medical reasons for admission. These results may help improve pre-admission motivational work with individuals, by basing it on their subjective experience.