Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Federici 2007

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

European Eating Disorders Review

Eur. Eat. Disorders Rev. 16, 1–10 (2008)

The Patient’s Account of Relapse


and Recovery in Anorexia Nervosa:
A Qualitative Studyy
Anita Federici 1* and Allan S. Kaplan 2
1
Department of Psychology, York University, Toronto, Canada
2
Loretta Anne Rogers Chair in Eating Disorders, Toronto General Hospital,
Vice Chairman Research and Director of Clinician Scientist Program, Professor
of Psychiatry, University of Toronto, Toronto, Canada

Objective: The purpose of this study was to explore the subjective


accounts of weight-recovered female patients, who met DSMIV
criteria for anorexia nervosa (AN), regarding their views of their
illness following weight restoration.
Method: Qualitative semi-structured interviews were adminis-
tered to 15 participants to ascertain their perspective of the factors
that either contributed to their maintaining a healthy weight, or the
factors involved in their having relapsed over the follow-up period.
Results: Qualitative analyses revealed six core categories: internal
motivation to change, recovery as a work in progress, the perceived
value of the treatment experience, developing supportive relation-
ships, awareness and tolerance of negative emotion and self-
validation.
Discussion: This study provides valuable information about the
way in which AN patients experience their illness and highlight
the factors that help or hinder recovery. These findings may help
enhance relapse prevention programs and potentially enhance our
ability to identify and target those individuals at the greatest risk of
relapsing. Copyright # 2007 John Wiley & Sons, Ltd and Eating
Disorders Association.

Keywords: anorexia nervosa; relapse; recovery; qualitative methodology

INTRODUCTION others continue to struggle, or relapse following a


period of remission, are unclear. Though longitudi-
The process of recovering from anorexia nervosa nal studies provide some evidence that relapse is
(AN) is a complicated and poorly understood more likely to occur among those who purge,
phenomenon. While intensive treatment may be present with a lower body weight, have a longer
initially successful in interrupting core symptoms, duration of illness, a later age of onset, and
long-term outcome of AN is variable. Specifically, significant comorbidity, the precise role of each
the reasons why some patients fully recover while determinant is not clear (Finfgeld, 2002; Herzog,
Schellberg, & Deter, 1997, Pike, 1998; Steinhausen,
* Correspondence to: Anita Federici, Department of Psychol- 2002). In their longitudinal study assessing outcome
ogy, York University, 297 Behavioural Sciences Building, 4700 in AN, Strober, Freeman, and Morrell (1997) found
Keele Street Toronto, Ontario, Canada M3J 1P3. significant associations between poor clinical out-
E-mail: federici@yorku.ca
y
This paper is based on the first author’s Master’s thesis in the
come and aversive family environments, compul-
Institute of Medical Science, School of Graduate Studies, sive desire to exercise at discharge and difficulties
University of Toronto. with socialisation. Carter, Blackmore, Sutandar-

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association.
Published online 26 July 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/erv.813
2 A. Federici and A. S. Kaplan

Pinnock, and Woodside (2004) identified several type of treatment and the perceived quality of the
additional predictors of relapse in AN, including therapeutic alliance play a significant role with
history of a suicide attempt, previous intensive regard to recovery and treatment satisfaction
treatment, obsessionality and concern about weight (Garrett, 1997; Noordenbos, Jacobs, & Hertzberger,
and shape at discharge. Furthermore, while the 1998; Rosenvinge & Klusmeier, 2000; Tozzi et al.,
greatest risk of relapse appears to occur during the 2003). Beresin et al. (1989) reported that treatment
first year post-treatment (Herzog et al., 1997), Carter compliance was enhanced when patients perceived
et al. (2004) found that subjects remained at an therapists as genuine, respectful, consistent and
increased risk of relapse up to 17 months post- non-judgmental.
intensive treatment, indicating that some patients The relevance and applicability of the above-
will relapse after remaining well for more than mentioned qualitative findings for AN patients in
1 year. the months immediately following completion of
Despite increased attention and awareness of both intensive treatment are not clear. Previous studies
the severity and chronicity of AN, significant gaps have typically interviewed patients several years
in our knowledge limit our ability to effectively treat (ranging from 6 months to 12 years) following
this disorder. Outcome measures and treatment various forms and durations of treatment. Given
responses have largely been defined by quantifiable that inpatient and day hospital programs are among
physical determinants (e.g. body weight), associ- the primary treatment interventions for individuals
ated physical sequelae (e.g. menses), attitudes and with AN, and in light of the greater potential for
beliefs about weight and shape, and measures of relapse immediately following treatment com-
comorbidity (e.g. personality, temperament). Such pletion, understanding the factors that facilitate or
approaches fail to take into account the view of the prevent change are paramount. To date, only one
patient in her response, or lack thereof, to treatment study has qualitatively explored relapse in individ-
and relapse (Garrett, 1997; Peters & Fallon, 1994). uals with AN following intensive treatment. Cockell
Unravelling the complexities of AN may well be et al. (2004) interviewed 32 female patients, the
aided by in-depth, qualitative studies which elicit majority of whom met diagnostic criteria for AN, 6
the patients’ view of treatment and reasons for months following completion of a residential
recovery and relapse. To date, only a small number treatment program. Participants identified a sense
of studies have methodically examined AN from of loss (e.g. lack of structure), feeling disconnected
this qualitative perspective. from others, engaging in self-defeating beliefs (e.g.
With regard to recovery in AN, qualitative studies unrealistic expectations about recovery, desire for
have consistently emphasised the importance of control) and difficulties dealing with the stress of
four key factors: social support, motivation for ‘real life’ (e.g. returning to work, daily hassles,
change, developing an identity independent from coping with the ‘diet culture’) as primary factors
the eating disorder and factors related to the that hindered recovery. Replication and extension
therapeutic alliance. Patients report that initiating of these data among AN patients in the months
and maintaining healthy and diverse, supportive following intensive treatment may have important
relationships (Beresin, Gordon, & Herzog, 1989; implications for relapse prevention programs.
Cockell, Zaitsoff, & Geller, 2004; Hsu, Crisp, & The aim of the current study was to investigate the
Callender, 1992; Pettersen & Rosenvinge, 2002; patients’ view of relapse and recovery utilising a
Tozzi, Sullivan, Fear, McKenzie, & Bulik, 2003) qualitative, phenomenological approach (Holloway
and/or removing oneself from destructive and & Wheeler, 2002; Morse & Field, 1995; Van Manen,
negative home environments, has served to facili- 1997). Specifically, we sought to identify how
tate adaptive change (Beresin et al., 1989; Hsu et al., patients conceptualise and understand their ability
1992). Willpower, personal strength, the desire for a or desire to maintain change in the first year
better life and psychological readiness have also following intensive treatment. In keeping with a
been identified as salient factors with regard to phenomenological approach (Creswell, 1998;
eliciting and maintaining change (Cockell et al., McLeod, 2001), every effort was made to ‘bracket’
2004; Hsu et al., 1992; Keski Rahkonen & Tozzi, preconceived ideas and assumptions about the
2005; Pettersen & Rosenvinge, 2002). Qualitative processes of recovery and relapse in AN. While both
studies have also shown that the ability and desire authors view recovery as a complex process that
to create an identity separate from issues involving involves multiple factors (e.g. nutritional rehabilita-
eating, weight and shape have been associated with tion, motivation, emotion regulation), the suspen-
recovery. Finally, patients with AN report that the sion of a priori hypotheses allowed the investigators

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
Relapse and Recovery in AN 3

to explore concepts as they were presented by the Structured Clinical Interview for DSM Disorders
participants without imposing biases or presuppo- (SCID; First, Spitzer, Gibbon, & Williams, 1996) and
sition onto the analysis. the Eating Disorder Examination (EDE; Fariburn
and Cooper, 1993) were administered. All inter-
views were conducted by the first author. Sub-
sequently, participants received one of two qual-
MATERIALS AND METHODS itative interviews, developed by the authors: the
first explored the participants’ perspective of the
Procedure
factors that contributed to their having maintained a
Patients with AN who had successfully completed healthy weight (BMI  19), having no bulimic
intensive treatment (BMI  20 kg/m2 at discharge) symptoms (binging, vomiting) and normal eating
at the Toronto General Hospital (TGH) were invited behaviours (n ¼ 7). The second interview assessed
to participate in the study. Briefly, participants had the participants’ view of the factors that contributed
completed inpatient or intensive outpatient treat- to their having lost weight and relapsed
ment. Treatment primarily focused on nutritional (BMI  17.5) following intensive treatment (n ¼ 8).
rehabilitation and utilised group-based cognitive Each interview began with an exploratory ques-
and behavioural interventions to target related tion which invited participants to describe which
psychosocial issues (e.g. interpersonal relation- factors, if any, they felt contributed to their (a)
ships, assertiveness, self-esteem, body image, sexu- having maintained a healthy weight or (b) having
ality). Following intensive treatment, all partici- lost weight and relapsing following intensive
pants participated in a manualised and controlled treatment. Participants were encouraged to speak
relapse prevention protocol that consisted of weekly openly about their perceptions and beliefs and were
outpatient CBT for up to 1 year. prompted occasionally by the interviewer to
All participants were female, 18 years of age or elucidate, develop and expand on their ideas. Once
older, and met diagnostic criteria for AN at participants had ample time to discuss their
admission to treatment as determined by a clinical thoughts, the interviewer proceeded to explore
psychologist. A total of 31 women were contacted additional content areas as follows: eating specific
approximately 1 year following discharge from factors (exploration of various factors related to the
treatment to assess their willingness and eligibility adoption or avoidance of weight control measures;
to participate in the current study. In order to i.e. caloric restriction, participation in exercise,
establish two clinically homogeneous groups, one vomiting, taking diet pills, etc.), environmental
clearly relapsed and one clearly recovered, partici- factors (family difficulties, extra-familial relation-
pants with a BMI  17.5 kg/m2 were considered ship difficulties, peer influences, cultural and
weight-relapsed and participants with a BMI  media, etc.), individual factors (personality traits,
19 kg/m2 were considered weight-recovered. Parti- presence and regulation of negative affect, self-
cipants whose BMI was between 17.5 and 19 kg/m2 esteem), genetic factors and family history. In
were not eligible for this study because recovery addition, interviews explored the patients’ view
status is more difficult to determine among patients of their treatment experience and how this may
with such body weights. Six patients did not meet have contributed to their having lost or maintained
the BMI criteria (BMI < 19 and >17.5 kg/m2). Two a healthy weight.
had a BMI greater than 19 kg/m2 but were actively
binging and vomiting. A total of 23 patients met the
Data Analysis
inclusion criteria. Five did not return our calls
following the initial screening, two stated that they In-depth interviews were tape recorded and
were no longer interested in participating in a transcribed verbatim into textual data. Qualitative
research study and one patient declined to partici- research software, NVIVO version 2.0, was used to
pate, as she was not willing to travel to the hospital aid in data organisation and analysis. Transcripts
for the interviews. A final sample of 15 participants were read in detail and a brief, interpretive
took part in the current study. summary was written capturing potential themes
All participants signed informed consent prior to and overall impression. Emergent ideas and
their participation in the study, which received thoughts were documented and this initial sum-
ethics approval from the TGH’s Research Ethics mary was discussed and explored with the thesis
Committee. In order to establish the clinical state of supervisor and thesis committee members. The
each participant at the time of the interviews, the language used by and the emotional presentation of

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
4 A. Federici and A. S. Kaplan

each participant were carefully considered through- t-recovered group were more likely to meet criteria
out the developing analysis. Significant statements for the restricting subtype of AN (85.7%) compared
pertaining to the phenomenon of interest were to those in the weight-relapsed group (50%), a
extracted and categories were developed. Names of Mann–Whitney U-test failed to show significance:
codes were created using the participants’ words. U ¼ 18.0; p ¼ .282. In terms of duration of treatment,
Similar codes were grouped together to form women in the weight-recovered group spent signi-
clusters of themes which were compared and ficantly less time in treatment (M ¼ 78.6 days,
contrasted with one another in order to verify SD ¼ 20.8) than those in the weight-relapsed group
common elements. Any discrepancies among and/ (M ¼ 116.8 days, SD ¼ 28.7; U ¼ 6.00, p ¼ .008). There
or between various clusters were re-analysed. The were no significant differences between the two
developing analysis was rated by independent groups in terms of treatment setting; in total, nine
reviewers in order to ensure credibility and validity. participants had been admitted to the inpatient unit
In addition, debriefing sessions with the supervisor and six had been day hospital patients. A two
and thesis committee were scheduled at regular (weight-recovered vs. weight-relapsed) by two
intervals to add credibility to the study and well as (presence or absence of a DSM IV Axis I disorder)
to ensure the integrity, rigor and dependability of contingency table was used to test differences
the findings. This allowed for an external check of between the two groups with respect to current
the research process whereby methodology, or lifetime presence of an Axis I Disorder as
interpretations and ideas were challenged, dis- measured by the SCID. Weight-relapsed partici-
cussed or reassessed. Finally, the results were pants were significantly more likely to meet criteria
integrated into an exhaustive description in order for a lifetime or current diagnosis of OCD (x2 ¼ 8.04,
to clearly illustrate the participant’s perspective of p ¼ .009).
relapse and recovery.
Qualitative Results
RESULTS Qualitative analyses yielded six core categories that
described the primary factors participants believed
Interviews were conducted on an average of contributed to their having lost or maintained their
14 months (SD ¼ 4.5) following discharge from weight following intensive treatment. Each category
intensive treatment. Each interview lasted approxi- is described in detail below:
mately 2 hours. Analyses revealed no significant
differences between eligible patients who agreed to
Category I: internal motivation for change
participate (n ¼ 15) and those who did not (n ¼ 8) in
Recovered Participant:
terms of admission BMI (M ¼ 15.8, SD ¼ 2.0 kg/m2),
duration of treatment (M ¼ 101.30 days, SD ¼ 28.9), I think that my own drive to get better and my
treatment setting (87% inpatient), prior number commitment to myself had been incredibly import-
of eating disorder hospitalisations (median ¼ 2, ant. I’m not doing it for anyone else because, as much
range ¼ 0–12) or lowest past weight (BMI; M ¼ as your parents and your friends care for you, it’s not
15.77, SD ¼ .43 kg/m2). There was a trend (x2 ¼ 3.63,
their problem. This is up to me. I have to do this.
p ¼ .089) for a greater percentage of those who
refused the study to be of the binge-purge subtype When asked to describe the factors that con-
of AN (75%) compared to those who participated tributed to their ability to maintain a healthy weight,
(33%). participants in the recovered group articulated a
At the time of the interviews, the mean BMI for the powerful desire to change, to let go of symptoms
15 individuals who participated in the study was and to move beyond the illness. Participants spoke
21.1 (SD ¼ 1.7 kg/m2) for those in the weigh- of being tired of their symptoms, of a desire to be
t-recovered group and 16.5 (SD ¼ 1.1 kg/m2) for healthy and of their recognition that the illness
those in the weight-relapsed group. The two groups conflicted with valued personal beliefs and
did not differ with regard to age (M ¼ 26 years, life goals. Participants stated that the decision to
SD ¼ 6.5), admission BMI (M ¼ 15.9, SD ¼ 1.94 kg/ recover had been a self-initiated and self-directed
m2), lowest past weight (BMI; M ¼ 15.9, SD ¼ process; rather than completing treatment for the
.50 kg/m2) or number of prior eating disorder sake of others, women in the weight-recovered
hospitalisations (median ¼ 2.3, range 0–12). Although group stated that they pursued recovery for
a greater percentage of participants in the weigh- themselves.

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
Relapse and Recovery in AN 5

Relapsed Participant: work was behind them. As a result, many stated


that they had been less strict about utilising
By the second week I knew it was a bunch of—and behavioural and self-care strategies in the months
I’d be sick the second I walked out of there, but I just following treatment. Unlike those in the recovered
had to show everyone that I could make it through group, participants who relapsed stated that they
the 12 weeks. I knew it wasn’t going to work! I was had not anticipated the challenges and struggles
like . . . I’ll humour them and put on the weight, but that presented themselves following intensive
I know . . . it won’t take me long to take it off again. treatment. Despite this, several weight-relapsed
participants did not see themselves as ‘relapsed’
Participants who had relapsed stated that they and argued that the investigator-imposed categor-
had been more ambivalent about their recovery and ies of relapse and recovery did not fully capture
their capacity to maintain change throughout, and their experience of their clinical state. Much like
following intensive treatment. They reported that those who had maintained their weight, weigh-
the benefits of using symptoms as a coping strategy t-relapsed participants considered themselves to be
continued to outweigh the costs. Many participants ‘in the process of recovery’.
reported that they had entered and completed
treatment in order to satisfy significant others. As a Category III: perceived value of the treatment
result, participants reported that they were ‘not experience
ready’ for treatment or that they had only been Recovered Participant:
‘willing to go so far’ in terms of symptom reduction.
I was with fellow women that suffered with the
Category II: recovery as a ‘work in progress’ same thing so that allowed me to be open about
Recovered Participant: issues . . . and because people were speaking about
them, I was able to speak about them.
. . . it’s not like, get out of the hospital weight
restored - you’re better! It totally was hard work, Women in the weight-recovered group reported
being on top of it, staying with the meal plan, keep that they were considerably more satisfied with
their treatment experience compared to those who
taking risks, being open and honest about things
relapsed. With regard to the intensive portion of
that are becoming scary again - because things do
therapy, weight-recovered participants stated that
become scary again! they felt safe, supported and validated by both staff
Participants in the weight-recovered group and fellow group members and that the specific
emphasised that recovery was an on-going process interventions (e.g. psychoeducation the structured
that required sustained vigilance and persistence. meal plan, weekly weigh-ins) had been essential
Even in one year post-treatment, participants did components of their recovery. Similarly, with
not view themselves as ‘recovered’, rather they respect to the relapse prevention phase of treatment,
considered themselves to be in the ‘process of participants who maintained their weight felt that
recovery’. Reaching weight-restoration in a treat- follow-up care had helped ease the transition out of
ment program was merely the first of a succession of intensive treatment and had been a crucial aspect of
steps. Many noted that their expectations about their ability to maintain change.
recovery had been fairly realistic and that they had Relapsed Participant:
anticipated and planned for the obstacles and
I feel like I didn’t deal with any of the issues that
challenges that presented themselves post-
were there to begin with. The bereavement, the
treatment.
Relapsed Participant: self-esteem issues . . . I hadn’t dealt with why I had
this eating disorder. How can I fix something when
. . . it was very hard to hang onto my weight. If there I don’t even know what the cause of it is?
was an opportunity to either eat safer or to cut out
In contrast, those who relapsed reported feeling
one or two exchanges, I wouldn’t hesitate . . .
dissatisfied with their treatment experience. Specifi-
Participants who subsequently relapsed reported cally, participants stated that the behavioural goals
that their experience of treatment and recovery of treatment (e.g. weight restoration) overshadowed
had been considerably different from their initial the identification and processing of core psycho-
expectations. Some reported feeling overconfident logical and emotional issues. As a result, and
about their recovery and assumed that the difficult regardless of having gained weight, participants

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
6 A. Federici and A. S. Kaplan

stated that they had not been well prepared for ‘life In contrast, those who relapsed reported that they
after treatment’. Many stated that they felt particu- did not have adequate social support and noted that
larly vulnerable and unskilled at managing ongoing they typically felt judged and misunderstood by the
emotional and interpersonal stressors in the months people in their lives. Marked by a pervasive sense of
following treatment. Likewise, participants repor- loneliness, isolation and alienation, participants
ted that follow-up care had been too inflexible and expressed frustration and disappointment that their
behavioural in orientation. They felt that too families and friends did not understand the
little attention had been given to issues involving difficulties they were experiencing. Many partici-
relationships, current stressors and negative emo- pants reported that their recovery was hindered
tions. As a result of not meeting their needs, many by the continual emphasis on body weight and
stated that they subsequently discontinued therapy. dieting within the family or social environment. In
Participants emphasised that either treatment nee- addition, participants stated that they had con-
ded to be longer or follow-up care more intensive to siderable difficulty accepting help from others. For
account for the difficulty they encountered trying to many, asking for help was perceived as a sign of
maintain change. weakness and failure.

Category V: awareness and tolerance of negative


Category IV: developing supportive relationships emotion
Recovered Participant:
Recovered Participant:
My family was very supportive at the beginning of
my recovery, and I think that was the most I am able to tolerate negative emotions and that has
important part . . . for them to take the illness had a very positive influence [on] stay(ing)
seriously . . . their understanding [and] willing[- well . . . on my getting recovered and maintaining
ness] to accept that I wasn’t deliberately doing this it. So, the negative emotions . . . [I’m] totally able to
to make myself sick. feel them and not get overwhelmed by them and not
have them translate into negative beha-
Participants reported that maintaining change
following intensive treatment was strongly related viours . . . that’s fundamentally important!
to the availability, support and non-judgemental Participants in the recovery group felt that
stance of family members and friends in the months learning to identify and tolerate negative emotions
following treatment. Participants emphasised the was central to their recovery process. Although
importance of feeling understood, accepted and of learning to sense and manage emotions began in the
being taken seriously by the people in their lives. hospital, the continual effort to work with emotions
Specifically, the adoption of a non-dieting approach in the months following treatment was considered
by significant others and spending time with people to be a significant contributing factor. Such work
who were less focused on body weight and shape, helped participants feel less overwhelmed and
were perceived to be significant factors that contri- threatened when negative feelings surfaced. Com-
buted to the recovery process. Moreover, partici- municating with parents, sharing feelings with
pants highlighted that recovery involved actively partners and talking with friends about their
‘letting people in’, asking for help and developing emotions increased their self-esteem and enabled
trust in others. For many, choosing to move towards them to develop a greater sense of control over
rather than away from people was a crucial part of negative affect. Participants stated that they gradu-
recovery. ally began to express their feelings in healthier,
Relapsed Participant: more adaptive ways.
Relapsed Participant:
I think I have trouble accepting help or accepting
that I might need help. I feel that it’s a weakness to I was feeling frustrated with having to feel
need help, like . . . I can do this on my own . . . just everything so I started drinking three days a week,
leave me alone. where it wasn’t just a little bit of drinking, it was
like . . . carry me home drinking . . . using the
I just felt very isolated and alienated from my alcohol to numb myself out. I needed something
family . . . they just didn’t know how to accept what else to take away the pain, to take away everything
I was going through . . .. that was inside that I wasn’t ready to let out.

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
Relapse and Recovery in AN 7

Those who relapsed reported that maintaining myself that I haven’t resolved, and until that totally
positive changes following weight restoration was resolves, eating is going to be an issue.
complicated by considerable difficulty tolerating
negative affect. Many stated that they had accom- In contrast, participants who relapsed reported
plished little during treatment in terms of identify- that their recovery was hindered by severe self-
ing and learning to tolerate unpleasant feelings. In criticism, intense fears of failure, a pervasive sense
the months following discharge, weight-relapsed of worthless and a feeling of hopeless for the future.
participants stated that they struggled to cope with They stated that they had considerable difficulty
intense feelings of depression, anxiety and lone- asserting their needs, acknowledging personal
liness. Furthermore, participants stated that the accomplishments and noted that they remained
discrepancy between their weight-restored bodies, heavily invested in the opinions, views and desires
which appeared ‘normal’, and their minds, which of others. Finally, participants reported that
continued to be consumed by ‘anorexic thoughts’, ongoing body weight and shape concerns played
was one of the most difficult issues they faced a significant role in their inability to maintain their
following treatment. Specifically, they reported that recovery.
it was increasingly difficult to relate to family and
friends who expected them to be free of their
disorder due an otherwise ‘healthy’ outward
appearance. In order to cope with their intense DISCUSSION
feelings, some participants reported that they found
In the current study, phenomenological method-
other, self-destructive ways of coping (e.g. self-
ology was used to discover how individuals with
harm, substance abuse).
AN understand the process of recovery and relapse
in the year following intensive treatment. This study
emphasises the importance of psychological change
Category VI: self-validation
in the recovery of AN. Though all participants
Recovered Participant:
responded behaviourally to treatment, in that each
. . . it’s two sides of the same coin. Self-discipline is was discharged from hospital after reaching their
target BMI of 20, the two groups differed in the
part of the reason I was able to be anorexic.
degree to which they were able to maintain their
Unfortunately you have to be extremely disciplined
recovery following treatment. The emergence of six
to starve for months and years, right? The flipside core categories illustrates that recovery was an
is the discipline is helping me stay with the eating, on-going process that required internal motivation,
not missing one single meal! belief in the capacity to change, the ability to
develop trust and the expression of emotion. In
Participants who maintained their weight
contrast, relapse was characterised by uncertainty,
reported that cultivating and maintaining a sense
reluctance to change, difficulty asking for help,
of personal worth and value was highly related to
intense negative emotion and a pervasive sense of
the maintenance of change following intensive
worthlessness. These findings have important
treatment. Participants described themselves as less
implications for relapse prevention and psychoe-
self-critical, more assertive, more accepting of their
ducation programs, as described below:
bodies and more likely to take credit for personal
successes and accomplishments. Maintaining 1. Motivation and recovery expectancies: These
recovery was strongly related to focusing on data expand our understanding of the role of
positive aspects of the self and challenging motivation in the processes of both relapse and
unhealthy thoughts about appearance and body recovery, and support the more recent emphasis
weight. Furthermore, many reported that person- on motivational interventions (i.e. motivational
ality traits once thought to have played an essential enhancement therapy) in the treatment of AN
role in the development of their disorder were now (Feld, Woodside, Kaplan, Olmsted, & Carter,
used to facilitate healthy change (e.g. patience, 2001; Tantillo, Nappa Bitter, & Adams, 2001;
stamina, determination, perfectionism). Treasure & Ward, 1997; Vitousek & Watson,
Relapsed Participant: 1998). They are also consistent with the work
of Hsu et al. (1992) and Pettersen and Rosenvinge
I think ultimately I got sick because of this (2002) who found that ‘being ready’ and feeling
incredible unhappiness and dissatisfaction with ‘motivated to change’ were salient factors in the

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
8 A. Federici and A. S. Kaplan

recovery process. Importantly, the current study relationships emerged as a significant factor that
found that, while motivation to change was cen- contributed to the maintenance of change follow-
tral during the initial stages of recovery, it was ing treatment. In addition to the presence or
also a key factor in later stages of recovery. absence of supportive relationships, those who
Participants who relapsed were more ambivalent maintained their recovery were differentiated
about change prior to, during, and after treat- from those who relapsed by three factors. First,
ment, whereas those who sustained a healthy social support was mediated by the ability and
body weight reported greater motivation to willingness to request and accept help. Many
change during and subsequent to intensive treat- participants who relapsed stated that asking
ment. These data also highlight the significance for additional help was weak, awkward or sha-
of recovery as an autonomous, self-motivated meful. Second, the adoption of a non-dieting
choice that was consistent with the long-term approach by family and friends was reported
goals and values of the individual. In addition, to have been a key factor that either facilitated
and consistent with the work of Cockell et al. or hindered recovery. Finally, the involvement in
(2004) and Constantino, Arnow, Blasey, and non-eating disorder-related community and
Agras (2005), expectancies regarding the recov- extracurricular activities appeared to support
ery process appeared to play an important role in the recovery process perhaps by fostering the
participants’ ability and desire to maintain growth of a new identity within the context of
change post-discharge. For example, compared a supportive, nurturing environment. These data
to those who relapsed, participants who main- suggest that post-treatment planning and relapse
tained their recovery stated that they had realistic prevention programs may facilitate treatment
expectations about continuing to work on symp- outcome by focusing specifically on enhancing
toms in the months following treatment. Internal problem-solving and interpersonal effectiveness
motivation, treatment expectancies and goal skills. Increasing patient awareness about the
identification may be important issues to empha- importance of social support and the obstacles
sise and measure in pre-treatment interventions involved in seeking help and encouraging
as well as in relapse prevention programs. patients to pursue non-eating disorder hobbies
Specifically, the inclusion of motivational strat- or activities would be beneficial. Additionally,
egies prior to, during, and after treatment may addressing the complexity of these issues in
serve to enhance therapeutic response and psychoeducation formats for family and friends
clinical outcome. would also be beneficial.
2. Emotion regulation: This study also highlights 4. Treatment and relapse prevention: The pub-
the importance of emotional processing and lished literature suggests that the quality of
affect regulation in the process of recovery from therapeutic alliance plays an important role in
AN. Based on these data, the ability to access, recovery and treatment satisfaction (Constan-
integrate and express emotions was central to the tino, Castonguay, & Schut, 2002; Horvath & Bedi,
maintenance of positive change following inten- 2002; Loeb et al., 2005). In the current study, the
sive treatment. Therapeutic approaches that degree to which participants felt a sense of safety,
place a greater emphasis on emotion regulation support and acceptance, by their therapists and
and exploration of internal affective states (e.g. fellow group members was reported to have
Dialectical Behaviour Therapy (DBT); Emotion- either helped or hindered recovery. These data
Focused Therapy) may have important implica- suggest that monitoring and placing a greater
tions for the treatment of AN. Recent findings emphasis on the development of the therapeutic
demonstrating positive outcomes using time- alliance over the course of intensive treatment
limited group-based and individual applications and throughout the relapse prevention phase
of DBT lend further support to the value of may improve treatment outcome for a greater
focusing specially on enhancing emotion regula- number of patients.
tion skills in the treatment of eating disorders Additionally, these data suggest that intensive
(Safer, Telch, & Agras, 2001; Telch, Agras, & programs may be sufficient interventions for
Linehan, 2001). weight restoration; however, for a subgroup of
3. Social support: Consistent with earlier studies patients, they do not foster lasting psychological
(Beresin et al., 1989; Hsu et al., 1992; Pettersen change. Given the emphasis on readiness and
& Rosenvinge, 2002; Tozzi et al., 2003), the pre- internal motivation for change, matching
sence of supportive and non-judgemental patients to treatment type and intensity and/or

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
Relapse and Recovery in AN 9

utilising stepped-care approaches may be phenomena of interest. Additionally, the women


particularly important for patients with AN. who participated in this study were all Caucasian,
Future studies evaluating the development and middle to upper class and well educated, which
application of individually-tailored treatments may further limit generalisability. Finally, partici-
would also be a valuable contribution to the pants did not read the transcripts or receive a
existing literature. While cognitive-behavioural summary of the results, which would have
interventions are particularly beneficial interven- enhanced the validity of the findings.
tions for patients with eating disorders, utilis-
ation of alternative treatment modalities may be
especially salient in the relapse prevention phase.
For example, in addition to focusing on weight
CONCLUSION
restoration, focusing explicitly on helping This study contributes valuable information about
patients develop comprehensive skills for nego- how patients with AN understand the processes of
tiating life stressors, interpersonal problems or relapse and recovery. This study demonstrates that
coping with emotion dysregulation may further AN patients have various and unique treatment
enhance treatment outcome. needs and that some patients may require different
5. Measuring and defining change: Although or additional treatment post-weight restoration in
additional research studies with larger sample order to facilitate long-term psychological change.
sizes are needed, these results have important Furthermore, this research highlights the import-
implications for the way in which we currently ance of using in-depth, qualitative approaches to
classify and measure treatment success and long- facilitate a richer understanding of the constructs
term outcome. Given that most participants, involved in the process of recovery that may be
regardless of their clinical state at the time of more difficult to access with traditional quantitative
the interviews, considered themselves to be ‘in methods. Clearly, there is a great deal to be learned
recovery’ highlights the complex and subjective from patients about their illness. Further qualitative
nature of change and progress. The inclusion of research in this area could contribute to the
more broadly based outcome measures and enhancement of relapse-prevention programs and
criteria that include behavioural as well as qual- facilitate the development of more individually
itative psychological measures (i.e. patients’ per- tailored and effective treatments.
ception of their illness and recovery) has the
potential to provide rich data with regard to
treatment response.
ACKNOWLEDGEMENTS
The authors wish to thank the patients who shared
Study Limitations
with us their personal views. We also appreciate the
While the use of qualitative methodology is a support and collaboration of Dr Marion Olmsted,
considerable strength of this study, it may also be a Dr Jacqueline Carter, Dr Blake Woodside, Dr Gary
limitation in that findings were based on the Rodin and Dr Mary Jane Esplen, who served on the
personal stories and information participants selec- first author’s Thesis Committee. In addition, the
tively chose to disclose. While every effort was authors would like to acknowledge Dr Pavla
made to build rapport between participant and Resnick who contributed to the conceptualisation
researcher, some participants may not have felt of the methodology.
comfortable revealing all the details of their
experiences. The study is further limited by its
retrospective design in that participants may not
have been able to clearly recall past events or REFERENCES
feelings. In addition, their recollection of the factors Beresin, E. V., Gordon, C., & Herzog, D. B. (1989). The
that contributed to their having maintained or lost process of recovering from anorexia nervosa. Journal of
weight may have been influenced by their current the American Academy of Psychoanalysis, 17, 103–130.
state of illness (e.g. whether a participant was Carter, J. C., Blackmore, E., Sutandar-Pinnock, K., &
Woodside, D. B. (2004). Relapse in anorexia nervosa:
weight-relapsed or weight-recovered at the time of A survival analysis. Psychological Medicine, 34, 1–9.
interview). Generalisability is limited by the small Cockell, S. J., Zaitsoff, S. L., & Geller, J. (2004). Maintaining
sample size and the fact that participants were change following eating disorder treatment. Pro-
self-selected based on their experience with the fessional Psychology: Research and Practice, 35, 527–534.

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv
10 A. Federici and A. S. Kaplan

Constantino, M. J., Arnow, B. A., Blasey, C., & Agras, W. S. psychotherapy (pp. 35–53). Thousand Oaks, California:
(2005). The association between patient characteristics Sage Publications.
and the therapeutic alliance in cognitive-behavioural Morse, J. M., & Field, P. A. (1995). Qualitative approaches.
and interpersonal therapy for bulimia nervosa. Journal In J. M. Morse, & P. A. Field (Eds.), Qualitative research
of Consulting and Clinical Psychology, 73, 203–211. methods for health professionals (2nd ed., pp. 151– 170).
Constantino, M. J., Castonguay, L. G., & Schut, A. J. (2002). Thousand Oaks, CA: Sage Publications.
The working alliance: A flagship for the scientist- Noordenbos, G., Jacobs, M. E., & Hertzberger, E. (1998).
practitioner model in psychotherapy. In G. S. Tryon Chronic eating disorders: The patients’ view of their
(Ed.), Counselling based on process research: Applying treatment history. Eating Disorders, 6, 217–223.
what we know (pp. 81–131). Boston: Allyn & Bacon. Pike, K. M. (1998). Long-term course of anorexia nervosa:
Creswell, J. W. (1998). Qualitative enquiry and research Response, relapse, remission and recovery. Clinical
design: Choosing among five traditions. Thousand Oaks, Psychology Review, 18, 447–475.
CA: Sage Publications, Inc. Peters, L., & Fallon, P. (1994). The journey of recovery:
Feld, R., Woodside, D. B., Kaplan, A. S., Olmsted, M. P., & Dimensions of change. In P. Fallon, M. A. Katzman,
Carter, J.C. (2001). Pre-treatment motivational enhan- & S. C. Wooley (Eds.), Feminist perspectives on
cement therapy for eating disorders: A pilot study. eating disorders (pp. 339–354). New York, NY: The
International Journal of Eating Disorders, 29, 393–400. Guilford Press.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. Pettersen, G., & Rosenvinge, J. H. (2002). Improvement
(1996). Structured clinical interview for DSM-IV axis I and recovery from eating disorders: A patient
disorders—patient version (SCID-I/P, version 2.0). perspective. Eating Disorders, 10, 61–71.
New York State Psychiatric Institute, New York. Rosenvinge, J. H., & Klusmeier, A. K. (2000). Treatment for
Fariburn, C. G., & Cooper, Z. (1993). The eating disorder eating disorders from a patient satisfaction perspect-
examination (12th ed.). In C. G. Fairburn, & G. T. ive: A Norwegian replication of a British study. Euro-
Wilson (Eds.), Binge eating: Nature, assessment and treat- pean Eating Disorders Review, 8, 293–300.
ment. New York: Guilford Press. Safer, D. L., Telch, C. F., & Agras, W. S. (2001). Dialectical
Finfgeld, D. L. (2002). Anorexia nervosa: Analysis of long- behaviour therapy for bulimia nervosa. American Jour-
term outcomes and clinical implications. Archives of nal of Psychiatry, 158, 632–634.
Psychiatric Nursing, 16, 176–186. Steinhausen, H. (2002). The outcome of anorexia nervosa
Garrett, C. (1997). Recovery from anorexia nervosa: A in the 20th century. American Journal of Psychiatry, 159,
sociological perspective. International Journal of Eating 1284–1293.
Disorders, 21, 261–272. Strober, M., Freeman, R., & Morrell, W. (1997). The long-
Herzog, W., Schellberg, D., & Deter, H. (1997). First term course of severe anorexia nervosa in adolescents:
recovery in anorexia nervosa patients in the long-term Survival analysis of recovery, relapse, and outcome
course: A discrete-time survival analysis. Journal of predictors over 10–15 years in a prospective study.
Counselling and Clinical Psychology, 65, 169–177. International Journal of Eating Disorders, 22, 339–
Holloway, I., & Wheeler, S. (2002). Phenomenology. In I. 360.
Holloway, & S. Wheeler (Eds.), Qualitative research in Tantillo, M., Nappa Bitter, C., & Adams, B. (2001). Enhan-
nursing (2nd ed., pp. 170–187). Oxford, UK: Blackwell cing readiness for eating disorder treatment: A rela-
Publishing. tional/motivational group model for change. Eating
Horvath, A. O., & Bedi, R. P. (2002). The alliance. In J. C. Disorders: The Journal of Treatment and Prevention, 9,
Norcross (Ed.), Psychotherapy relationships that work: 203–216.
Therapists contributions and responsiveness to patients Telch, C. F., Agras, W. S., & Linehan, M. M. (2001).
(pp. 37–69). New York: Oxford University Press. Dialectical behaviour therapy for binge eating dis-
Hsu, L. K., Crisp, A. H., & Callender, J. S. (1992). Recovery order. Journal of Consulting and Clinical Psychology,
in anorexia nervosa: The patient’s perspective. Inter- 69, 1061–1065.
national Journal o f Eating Disorders, 11, 341–350. Treasure, J., & Ward, A. (1997). A practical guide to the use
Keski Rahkonen, A., & Tozzi, F. (2005). The process of of motivational interviewing in anorexia nervosa.
recovery in eating disorder sufferers’ own words: An European Eating Disorders Review, 5, 102–144.
internet-based study. International Journal of Eating Tozzi, F., Sullivan, P. F., Fear, J. L., McKenzie, J., & Bulik,
Disorders, 37, S80–S86. C. M. (2003). Causes and recovery in anorexia nervosa:
Loeb, K. L., Wilson, G. T., Labouvie, E., Pratt, E. M., The patient’s perspective. International Journal of Eating
Hayaki, J., Walsh, B. T., et al. (2005). Therapeutic Disorders, 33, 143–154.
alliance and treatment adherence in two interventions Van Manen, M. (1997). Researching lived experience: A
for bulimia nervosa: A study of process and outcome. human science for an action sensitive pedagogy
Journal of Consulting and Clinical Psychology, 73, (2nd ed.). Ontario, Canada: Althouse Press.
1097–1107. Vitousek, K., & Watson, S. (1998). Enhancing motivation
McLeod, J. (2001). The phenomenological approach. In J. for change in treatment resistant eating disorders.
McLeod (Ed.), Qualitative research in counselling and Clinical Psychology Review, 18, 391–420.

Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association. Eur. Eat. Disorders Rev. 16, 1–10 (2008)
DOI: 10.1002/erv

You might also like