Anorexia nervosa is one of the most common chronic medical conditions among teenage girls-one source ranks it third, after obesity and asthma1-and it's on the rise in that population.2, 3 Prevalence among women and girls, who constitute 90% of all cases, is estimated to be as high as 0.5%.2, 4

It's commonly believed that medically compromised children and adolescents with severe anorexia benefit from treatment in specialized pediatric acute care settings.5 The majority of Australian pediatric inpatient programs are based on behavior modification principles that promote nutritional and medical stability through refeeding. Additional motivational or family-oriented therapies and psychological interventions may be initiated in the hospital, but they are conducted primarily as outpatient services following discharge.6 Similarly, in Europe and the United States, severe cases may be treated within behavioral refeeding programs in either inpatient settings, which provide 24-hour access to medical and nursing staff, or intensive residential centers, which provide rigorous, structured treatment in a homelike environment (for patients who are medically stable).7

Nurses play a crucial role in such programs, providing 24-hour care, support, encouragement, and guidance through the establishment of a trusting therapeutic relationship, which is widely cited as being vital to the successful treatment and recovery of people with anorexia.8-10 Nevertheless, there's been relatively little research published on the difficulties nurses face in forming therapeutic relationships with adolescents who have anorexia or on the effects inpatient hospitalization and behavior-modification treatment have on the therapeutic relationship.11

This study examined an Australian inpatient behavioral program for adolescents with anorexia. The program is conducted within a 20-bed, general hospital ward (what U.S. nurses would call a "unit") for adolescents with a variety of medical or surgical conditions. Typically, four beds in the ward are assigned to adolescents with anorexia, but during the write-up of this study, the number of beds allocated to such patients was increased to six. Based on medical and psychological assessment, patients enrolled in the program are assigned to one of four progressive treatment levels (see Levels of Treatment).

The investigator (LMR) conducted in-depth, face-to-face, semistructured interviews with 10 adolescents who were treated in the program and 10 of the pediatric nurses who cared for them. The results suggest that both nurses and patients face difficulties in forming therapeutic relationships within such an environment. In recounting their experiences, interviewees often used language that showed they thought the program functioned as a metaphoric prison, with nurses taking on the role of prison warden. All participants (patients and nurses) tended to concentrate on three major phases of life within the program, which were thematically categorized as "entering the system," "'doing time' within the system," and "on parole or release." We hope that these study findings, and the recommendations for change they suggest, may improve therapeutic relationships within such programs.

Levels of Treatment

Patients enrolled in the adolescent anorexia treatment program that is the subject of this study are assigned a treatment level based on a medical and psychological assessment. Patients are admitted on treatment level one or two; as they progress toward medical stability, gain weight, demonstrate improved psychological functioning, and adhere to program regulations, they graduate to higher treatment levels at which they are given an increasing number of privileges. The program's four treatment levels are as follows:

* Level one is reserved for medically unstable patients who require complete bed rest (occasionally with toilet privileges); receive a continuous, high-calorie nasogastric feed; are closely monitored; and are only allowed visits from immediate family members (for up to two hours daily on weekdays and three hours daily on weekends).

* On level two, patients eat supervised meals in the dining room, followed by 30 minutes of supervised rest on lounge chairs. They attend school and group sessions, begin physiotherapy, and are allowed one 10-minute shower per day. Visitation is restricted, as it is on level one. Although less restrictive than level one, a level two designation does not allow patients to leave the ward at any time.

* On levels three and four, patients are permitted to leave the ward (but not necessarily the hospital) under supervision for 20 and 40 minutes per day, respectively. Visitation hours remain the same, but on levels three and four, patients are allowed visits from one additional relative (outside of the immediate family). Patients' weight and progress (measured in terms of adherence to the program and improvement in psychological health) are evaluated twice a week. Progress may be rewarded with gate passes, which (depending on the degree of progress) permit the patient to leave the ward with family for either one meal and one snack (half-day pass) or two meals and two snacks (full-day pass), either within the hospital or off hospital grounds; and for patients on level three, advancement to the fourth level.

METHODS

To explore how patients and nurses experience the program, the investigator employed naturalistic inquiry, an interpretive research methodology conducted within a constructivist paradigm.12, 13 Purposive sampling was used to ensure adequate representation across the spectrum of nurses working on the ward (in terms of their experience working with adolescent anorexia as well as credentials) and patients treated on the ward (in terms of number of admissions and days per admission). RNs and enrolled nurses with more than one year's experience in the program were recruited (in Australia, enrolled nurses are those who complete an 18-month-to-two-year diploma course that enables them to perform specific duties under the supervision of an RN), and those who spent the greatest amount of time working with adolescents in the anorexia program were interviewed first. Four previous patients were invited to participate in the study because they had been admitted multiple times and had spent the most time as inpatients.

Interviews were conducted up until the point of saturation (when responses ceased to reveal any new information or themes). Nurses were interviewed twice, with one exception; all patients were interviewed once. Patients who were medically or cognitively unstable and those determined by their treating physician to have severe psychiatric comorbidities were excluded from the study. The process yielded a total of 29 interviews: 19 with nurses (eight RNs and two enrolled nurses) and 10 with patients, six of whom were inpatients during the study period (June through December of 2004) and four of whom were previous inpatients. Five of the patients had undergone multiple readmissions (see Tables 1 and 2).

Table 1. Profiles of Adolescent Participants

Table 2. Profiles of Nurse Participants

This particular program was selected for study because, at the time of the study, the investigator worked part-time at the site and had an in-depth understanding of the ward routine and practices. To minimize the potential for overidentification with nurse respondents, co-constructions were validated through reflexive diary writing, peer debriefing, and member checking.13 The hermeneutic-dialectic approach to data analysis requires researchers to explore both convergent and divergent viewpoints to verify that constructions accurately represent the discourse within the time frame and context of the study.14

The data collected were audio recordings of 1.5-to-2.5-hour semistructured interviews conducted over seven months in a private room on the ward. Because responses often prompt spontaneous questions, the interview structure was flexible, based on a general guideline that focused on six broad topics, rather than on a strict set of specific questions (see Interview Guideline). Each interview began with the open question, "Can you please tell me about the nursing care for adolescents with anorexia on this ward?"

Box. Interview Guideline

Data collection and analysis occurred concurrently, using a method known as constant comparative analysis.13, 14 The audiotapes were transcribed verbatim, read, reread, and then were subjected to thematic analysis both manually and with the qualitative data analysis software program NVivo 2. Themes communicating the essence of the ideas expressed within each interview were identified, and the participants checked the themes.

Ethical approval was obtained from the children's hospital in which the study was conducted and the University of Western Sydney in New South Wales, Australia, where the investigator was completing a doctoral program in nursing. All participants (and parents of adolescent participants) received both an oral and a printed explanation of the study and provided, in turn, written consent to participate. Participation was voluntary, and participants were allowed to withdraw from the study or terminate an interview at any time (none did so). In data compilation, participants were identified only by pseudonyms.

RESULTS

In qualitative research, metaphors often function to "illuminate the meanings of experiences."15 In this study, the language and terminology instinctively adopted by the participants, particularly the adolescents, provided insight into their lives on the ward-both nurses and patients consistently used the metaphor of prison life to articulate their experiences. The prison metaphor thus provided a framework by which the data could be interpreted and presented. Three major themes emerged from the interviews: "entering the system," "'doing time' within the system," and "on parole or release." These major themes, and corresponding subthemes, reflected the similarities between life on the ward and life behind bars.

Entering the system. Adolescents entered the system in one of two ways. Either they were taken to the ED by a concerned family member, or they were attending a clinic appointment when the decision was made to admit them. Megan described her first admission as a "terrible, traumatic" experience. Others recalled many emotions, including fear, anger, depression, and confusion, about why they were being admitted.

Officer, what's the offense? Danielle "never thought that someone could come into hospital for that kind of condition," and it made her think "I shouldn't be in here." Megan thought she was "en route for a holiday" when her family suddenly admitted her for treatment. As she recalls the day: "I didn't even know we were stopping at the hospital. We were stopping in for counseling or something. I didn't know[horizontal ellipsis]. Then I found out straightaway that I was being admitted and my parents had to leave within [horizontal ellipsis] half an hour of dropping me off."

In retrospect, Cameron and Melinda believed that their admission "was needed," and Danielle agreed, saying that "finding out about all the harmful sides that this illness can do to you, then you start to realize, 'Well, I should be in here.'"

Locked up. Nurse Donna remarked that an adolescent may view level one as "being locked up in jail [horizontal ellipsis] isolated from everyone else." Cameron corroborated this observation, noting that being unable to see friends or go outside made him feel "quite cooped up," like being "in jail in a way."

The program limited normal activities such as going to the bathroom and leaving the ward. These became privileges, to be earned by adhering to the rules and moving through the levels. With each level progression, patients saw themselves as increasing their chances of making their "way out the door."

In general, nurses believed the program's intentions were "honorable" and that they had a duty to follow the program. Nurses unanimously perceived the adolescents as not adhering to socially acceptable standards, as not having "what we would class as a normal eating habit or a normal view of food." Most nurses saw their role as helping the adolescents return to normal eating patterns of "three square meals a day" and described the program's goal as "changing or altering a particular behavior [horizontal ellipsis] that's self-harming [horizontal ellipsis] deliberate or obvious [horizontal ellipsis] into something that's healthier." Nurses saw correcting adolescents' eating habits and behaviors as a means of reintegrating them into society "so they are not so separated from [horizontal ellipsis] other teens." Behavior modification was the tool for eradicating anorexia, the undesired behavior.

The aim was to "identify what's causing the child to behave in this way" and to "get them physically better, as well as mentally better." The nurses viewed the program in a positive light overall, though they questioned the adequacy of the psychological therapies provided, noting that, with its rewards and punishments, the program emphasized the physical aspects of getting the patients "medically safe to be at home," while doing little to prepare them psychologically.

All adolescents within the program described their daily routine in the same way-as highly structured, with little or no variability, much like that of a prison-unlike that of adolescents on the ward who were receiving treatment for other conditions. While the other adolescents had freedom and flexibility, those in the anorexia program had a fixed and constant schedule, much like journalist Morgan describes that of prison inmates, who "eat, rest, and work" when ordered, with "little choice but to accept the rules which regulate their lives."16

Patients without an eating disorder were allowed to see unlimited friends and relatives during visiting hours; on weekends, they could sleep in and have a late breakfast. By contrast, adolescents with anorexia were not permitted to have nonrelatives visit, and they were required to remain faithful to their schedule throughout the week. Their day was set like clockwork, with assigned times for waking, showering, eating, sleeping, attending school, participating in group activities, going to motivational or family therapy (if that was included in their individual treatment plan), and seeing visitors. Free time was limited to four and a half hours per weekday and about nine hours per weekend day.

Nurses explained that certain small concessions made the experience of the program less oppressive. Patients were "allowed to decorate" their personal bed spaces and were "not forced to [horizontal ellipsis] wear hospital pajamas." They were "allowed to be individuals" in the way they dressed. Nurse Thomas described the ward as a second home for the patients, noting that many tried to "brighten up the place" and make it more comforting by bringing in their own bed quilts and decorating their bedsides with "arts and crafts [horizontal ellipsis] pictures [horizontal ellipsis] things they have painted [horizontal ellipsis] cards from school friends." As Hutton explains in her report on the use of ward space by adolescents, the creation of unique, personal spaces enables adolescents to escape "the designated spaces of being a patient, a body, an illness, a part of a routine" and to express their identity.17

Learning the ropes. Once over the initial shock of entering the program, adolescents familiarized themselves with the routine and learned the ropes from fellow inpatients. Success was measured by "getting better" and "getting out." Actions and behavior would determine how long each remained an inpatient.

On arrival, each adolescent was given a copy of the program, which included the ward timetable and an explanation of program rules and expectations. Mealtime rules were discussed by all adolescents interviewed, with particular emphasis given to what they were not allowed to do. As Sara explained, "We wouldn't be allowed to cut bread into [horizontal ellipsis] six or eight pieces [horizontal ellipsis] and leave crumbs [horizontal ellipsis] and wipe the butter on the napkin and things like that."

After main meals, the adolescents described how they were required to sit on lounge chairs positioned near the nurses' station. Nurse Donna explained that this was "so all the nurses can keep an eye on them." Zoe explained that first-timers used this rest time to learn more about the program: "If there was a new girl in, she would always ask questions then."

'Doing time' within the system. In prison slang, the phrase "doing time" has been described as "the sense of futility and waste that is a prison sentence."18 A few of the adolescents in this program saw their multiple, lengthy admissions in this light. The realization that they didn't want to spend the rest of their lives in a hospital was an impetus for recovery. Similarly, prison inmates often describe their prison sentence as "a 'pause' in their lives."19

The first time many of the adolescents in this program discovered that they would be doing time in a hospital, they were in "shock," "depressed," "angry," "on the brink of tears," and most of all, "scared" about the thought of hospitalization. Frustration over the program's restrictive nature was also a common initial reaction.

For example, Josephine explained that, on admission, she wasn't allowed toilet privileges, though "the bathroom was right next to [me]!" The time spent on bed rest was "boring" for many who remained on this restriction for as long as a week. Cameron remembered the days spent on bed rest as consisting of "nurses coming in all the time checking your blood pressure and your pulse and giving you your meds and everything like that." He found it difficult because he was "so used to [horizontal ellipsis] running around and stuff." Amber, on the other hand, didn't seem to mind that she "wasn't [horizontal ellipsis] allowed to do anything that required [her] leaving bed." She recalls being "pretty sick" when she came in and having "no energy." Most patients explained that their initial negative reactions to hospitalization was because they didn't fully understand the complexity of their situation.

Danielle felt that being hospitalized became easier with each subsequent admission because she knew the environment, the program, and the nurses. "You feel a bit more comfortable than when you first came in," she said. Megan had the opposite reaction, finding subsequent admissions more difficult and feeling a greater need to rebel against the restrictions. "It got tougher as it went along," she explained. "First time it was kinda easier [horizontal ellipsis] because I didn't really know what was happening[horizontal ellipsis]. Then the more times I was admitted [horizontal ellipsis] the harder it got because I wanted to be more independent." Commonalities can be seen between these patients and prison inmates, with some settling into the new environment easily ("doing it easy" in prison slang), and others unable to accept the "sentence" ("doing it hard").18

Surveillance. In discussing the intensive monitoring the patients required, Nurse Paige summed up the feelings of most of her peers, saying that it seemed like the adolescents were under constant "surveillance." A primary responsibility of nurses in the program was to enforce rules and regulations to which the adolescents were not particularly receptive. While this monitoring was crucial to ensuring patient safety, it left many of the nurses feeling drained and in a constant state of "watching and wondering" what their patients would be up to next.

All the adolescents bemoaned that the program didn't allow visits from friends. As Isabel remarked, "Well, at first I didn't really like the fact that [horizontal ellipsis] friends couldn't come and visit [horizontal ellipsis] I think they should be able to[horizontal ellipsis]. It doesn't really make sense to me." Echoing Isabel's frustration and reflecting the prison metaphor, Josephine commented that it was "highly unlikely" that friends would "scheme something to get [us] out."

The adolescents believed that visits from friends would help to "get [their] mind[s] off things" and keep them "in contact" with the outside world. Zoe said she could see the positive effects that friends had on other patients in the ward, "making them feel so much better." She also pointed out that "if you are not allowed to see your friends and then you go back to school and no one has seen you for [horizontal ellipsis] two months [horizontal ellipsis] it's a shock for everybody." These comments demonstrated a sense of detachment from the outside world, analogous to the detachment some prison inmates reportedly feel when released.20

The nature of the disease required the program to limit such personal liberties as toilet and shower use. Showers, monitored by nurses, were allowed in the morning and "shouldn't be more than ten minutes." Nurse Donna explained that the patients "have free use of the toilet as long as it is not straight after main meals[horizontal ellipsis]. If they are a bit long, I [horizontal ellipsis] listen at the door and make sure they are not vomiting or anything."

Excessive bathroom use would result in the adolescent being placed on "toilet restrictions." Normally, this would mean a maximum of two visits to the bathroom per eight-hour shift, and "they have to ask [nurses'] permission to go." Some nurses were strict with these rules for those who had "been caught" vomiting, purging, or exercising in the bathroom. As Nurse Gabrielle reported, "I would see no problem in giving them a bedpan." Similarly, excessive exercising by adolescents resulted in their being placed on 24-hour bed rest; adolescents who were nonadherent, left the hospital without permission, or engaged in self-harm were "specialled"-placed in a single room on the ward and kept on level-one treatment under 24-hour watch by a nurse. Zoe, one of the discharged patients interviewed for the study, described a time when she had been on "nurse watch." She explained that nonadherence with the program "discredits your whole reputation and so the treatment you get is often a lot more concentrated." Such "special" treatment is in many ways comparable to seclusion, which is often used in prisons to deal with a troublesome or suicidal inmate,21, 22 though the "specialled" patients do interact with nurses and other clinicians in the program.

No more privacy. The nurses acknowledged that anyone coming into a hospital would experience a certain loss of privacy, which was "one of the biggest drawbacks" to hospitalization. For the adolescents in the anorexia program, this infringement would "probably be slightly more than [for] a normal patient, given their condition." As Melinda described her experience in the program: "You don't really have very much privacy, and they're always wanting to know what you're doing and why you're doing it." Nurses said that they would "bust in on [the adolescents] in the bathroom" if they felt the young people were "breaking the rules." They'd also go through patient belongings and perform "locker checkups" if they believed patients were "storing supplements," such as laxatives, or consuming prohibited food items, such as sugar-free chewing gum or soft drinks. Nurses even mentioned having "to check [that there was] no syringe placed in the rubbish bin" in patients' rooms, because syringes could be used to withdraw feeds from their nasogastric tubes. The locker and room searches nurses described are reminiscent of the cell and strip searches conducted in prisons when prisoners are suspected of concealing contraband.

Zoe and Josephine, two of the discharged patients, reported feeling that their privacy had been invaded when they discovered that their personal belongings had been searched. As Zoe said, the fact that nurses had gone "through my stuff when I wasn't there [horizontal ellipsis] really shitted me off." Nurses, too, disliked the monitoring and invasion of patient privacy, describing these as "the parts [of the job they] hated most." To implement the program successfully, however, it was necessary for them to monitor eating, weight, and bathroom activities and to administer consequences for inappropriate behavior. Megan acknowledged that the invasion of her privacy was an annoying aspect of the program, but she realized that it was part of the nurses' job and that "they only do it for your own good."

Time off for good behavior. Twice a week, before breakfast, patients-wearing only their underwear and a hospital gown-were weighed after voiding in a receptacle. Patients were not allowed to void in the toilet to ensure that they didn't "water load" (drink lots of water while in the bathroom to falsely boost their weight).

The adolescents described the stress of waiting for the physicians to hand down their verdicts-whether they would be rewarded or penalized for their efforts. For most, the stress of a weigh day was twofold: putting on weight was good in that it usually meant a "chance to move up a level," but bad because it made them feel that they were getting "fat." Losing weight meant losing privileges as well. As Amber explained, "If we lose weight, they find some sort of way to punish us-put our meal plan up, move us back down a level, or whatever." From Megan's perspective, it was particularly hurtful to be refused a gate pass if her weight remained stable, because gate passes were something the patients really "look[ed] forward to." They represented "freedom," an opportunity to go home, or go out shopping with friends or family, and a chance to return to "normality." Nurses also saw gate passes as an important "reward"-an opportunity for patients to get some fresh air, exercise, and escape from their constant watch.

Life as prison wardens. In relation to their patients, nurses working in this program functioned much like prison wardens overseeing inmates. Nurse Oliver described it as "shut[ting] down [horizontal ellipsis] go[ing] into [horizontal ellipsis] policeman mode," which he felt nurses did in "self defense," to avoid cracking under the strain of the job.

Given the program's behavioral orientation, it operated largely by using rewards and punishments. Nurse Oliver likened the program's behavioral modification principles to the methods used in "training dogs." Although he identified both negative and positive reinforcements as important, he emphasized the influence of the negative, explaining that "the threat of [a nasogastric] tube is a negative reinforcement to bad behavior [and] can encourage them to behave well [horizontal ellipsis] push themselves harder to comply with what they know is expected of them[horizontal ellipsis]. They want to go home because being in hospital is not a particularly pleasant experience [horizontal ellipsis] they want to have gate passes on the weekends[horizontal ellipsis]. That's all positive but [horizontal ellipsis] the actual being here is a negative in itself."

Caring for the adolescent patients in this program became "very routine" and "monotonous" for most nurses. Nearly all saw themselves go into "autopilot" on the ward because they "[knew] the routine inside out." All the nursing tasks-checking meal trays, taking vital signs, weighing and testing urine samples, ensuring adherence to bed rest or school attendance-became "habitual." There was no variation from the routine except on weekends. In large part, the monotony grew out of the difficulty nurses faced in establishing therapeutic relationships with the patients. As nurse Gabrielle explained, "You do what you have to do, and you get out. And [horizontal ellipsis] that is partly because it is such a challenge [horizontal ellipsis] to interact with [the patients] or get them to interact with you and [to] build a rapport[horizontal ellipsis]. Because you're doing things to them that they are against."

On parole or release. The adolescents realized that to be released they needed to adhere to program regulations and achieve their goal weight. Danielle described recovery as "getting back into a normal life without [an] eating disorder." She viewed it as a journey along "quite a long road [horizontal ellipsis] which [horizontal ellipsis] [has] ups and downs before you actually reach the end where you're completely recovered." Although there were many "hurdles" to conquer, Zoe felt it was "better being out than in."

When adolescents were discharged, they were expected to return to the hospital's adolescent medicine unit for weekly "checkups" with physicians, who assessed their ability to cope in society while maintaining a healthy weight and social life. Just as parole violations may require a prisoner to return to prison, the adolescent's inability to maintain a healthy weight would necessitate readmission to the anorexia treatment program.

Relapse rates are high among adolescents with anorexia,23 as are recidivism rates within Australian prisons.22, 24 In this program, many of the adolescents were readmitted several times. For patients who returned, Nurse Zac measured recovery in terms of "the length of time between admissions," adding that "sometimes they [horizontal ellipsis] fear going back into society." Nurse Mandy reiterated Zac's sentiment, remarking that "here they feel safe" and noting that some "played up" prior to discharge-by not adhering to meal requirements, for example-in order to be put back on a lower level. Danielle, a patient, agreed, explaining that "some girls don't want to go home. They feel that they're not ready." Similarly, prisoner recidivism is often attributed to inadequate preparation for reintegration into society and the lack of support within the community following release.22, 24

DISCUSSION

Thematic analysis revealed that the challenge of forming positive therapeutic relationships is magnified for nurses working with adolescents being treated for anorexia in an inpatient behavior modification program. In such programs, the ward may function as a metaphoric prison, with patients seeing themselves as inmates and nurses as prison wardens.

Both prisoners and adolescents with anorexia are institutionalized because their behavior has been deemed deviant. As nurses in this study related, they viewed the behaviors associated with anorexia as deviant and dutifully imposed sanctions in an attempt to modify or reduce those behaviors, much as the justice system imposes sanctions on those accused of criminal behaviors. People who are labeled deviant often "seek to resist or ignore that interpretation of them and their behaviour."25 And if they happen to be patients resisting the interpretations of nurses, the development of a therapeutic relationship may be in jeopardy.

Psychiatry supports the use of behavior modification principles as a means of altering dangerous eating behaviors, thereby promoting rehabilitation and positive changes in a person's health and well-being. Similarly, prisons "are designed primarily to contain" the prison inmates, "but also to correct or rehabilitate" their deviant behavior.22, 26 The parallels between the penal system and the inpatient program for adolescents with anorexia that was the focus of this study are apparent. The adolescents in this program were not free to discharge themselves. For all practical purposes, they were locked up, deprived of personal liberties and social contacts, until they gained weight and achieved medical stability. But like the adults with eating disorders in the study by Guarda and colleagues,27 the adolescents in this program eventually came to accept help, although they initially denied needing it.

The inherent conflict between administering treatment based on behavior modification, on the one hand, and developing therapeutic relationships, on the other, may pose the greatest challenge for nurses in this type of program. With the intense demands on their time, nurses within this acute care program reverted to task-oriented actions, becoming protocol-driven and tending to diminish the significance of caring and the importance of the psychosocial aspects of care.

Nursing philosophy professes a holistic model of care that is critical of task-oriented care and often conflicts with the narrower biomedical model exemplified by behavior modification programs.28 In many ways, nursing care for the adolescent with anorexia reverted to a type of custodial care, being "a passive watcher and guardian,"29 with nurses' attempts to establish a therapeutic relationship limited by the role they played as enforcers.

Nurses' relationships with adolescents in this program were influenced by the role they were assigned, by their fundamental mistrust of their charges, and by the program's rigid rules and protocols. Many nurses principally attended to the physical care tasks, perhaps as a means of avoiding the other dimensions, which might be more emotionally challenging and frustrating.

Because adolescence itself is a period in which issues of rebellion, power, authority, and identity come to the forefront, and because anorexia is a disorder that revolves around power and control over eating, weight, and exercise, using behavior modification principles in the treatment of adolescent anorexia is challenging. The adolescents, not unlike prison inmates, were confined and dependent on others when held within the system. Their outlet was to rebel so as to recapture something of their independence. Manipulating staff (through dishonesty, flattery, or by pitting one staff member against another) and breaking rules (concealing heavy objects in their clothes on weigh days, throwing away food in napkins, or sabotaging nasogastric feeds) allowed them to maintain a stronger hold on their eating disorder while rebelling against the program's authority figures.

Building relationships with the adolescents was particularly difficult for nursing staff, who were required on a daily basis to invade the adolescents' privacy, supervise their personal activities, and monitor manipulative behavior. While other studies stress the challenge of forming therapeutic relationships with adolescents with anorexia,11, 30 our analysis demonstrates that it's particularly challenging within the context of a behavior modification program. From the nurses' perspective, following the program meant they were doing something ethical, that is, saving a life. Yet their actions in doing so diminished patient trust and limited their capacity to form therapeutic relationships with their patients.

Just as prisons see high recidivism rates,22, 24 anorexia treatment programs see high relapse rates and become a revolving door for some of the adolescent patients.23 That was true of this program as well.

Limitations. Because this study was confined to one adolescent ward in an acute care pediatric setting and included such a small sample of patients and nurses, it is difficult to generalize its findings to other settings. The study is further limited by the fact that the investigator worked part-time at the site at the time of the study and may have been inclined to over-identify with nurse respondents.

CONCLUSION

The prison metaphor that informs relationships between nurses and patients on this ward clearly needs to be examined, revised, and replaced. Reflecting upon the dynamics of the metaphor may help nurses better understand how it shapes the feelings and behavior of adolescents in the program and impedes the development of therapeutic relationships.

Given the difficulties inherent in inpatient anorexia treatment, programs such as this one should strive to limit hospitalization to the time required to establish medical stability, so that patients return more quickly to the community, where they can enter primary counseling under circumstances more conducive to a successful therapeutic relationship. This is not to say that nurses working in inpatient anorexia treatment programs shouldn't attempt to establish therapeutic relationships with their patients. On the contrary, Tan and colleagues validated the importance of a trusting therapeutic relationship and a supportive environment as empowering patients with anorexia to be active recipients of care.31 They acknowledge, however, that such adolescents struggle with issues of control, particularly within a behavioral program, and that this struggle may engender resistance. Providing adolescents with an opportunity to actively participate in some treatment decisions-perhaps through contracts that outline patient/nurse responsibilities or clarify confidentiality rules, so that patients know the type of information they may disclose to nurses without fear that it will be passed on to physicians or documented in the medical notes-may promote the formation of therapeutic relationships. As nurses in Dearing's study of schizophrenia treatment expressed, the power of a therapeutic relationship lies in "truly caring for [patients] as people, not seeing them as an illness, recognizing the uniqueness of them, and giving them a sense of [horizontal ellipsis] control over their own lives. It's letting them know they are in charge of their recovery."32

Implications for practice. Some comments from nurses, such as Oliver, who likened the program's behavioral methods to dog-training strategies, suggest that a better understanding of behavior modification principles may help nurses focus more on promoting positive behaviors in adolescent patients, rather than on eradicating negative behaviors. Mental health education and training in counseling would likely increase nurses' confidence in dealing with fragile adolescents. Likewise, knowledge of anorexia and the distress it causes patients may improve nurses' relationships with affected patients.9

Our patient interviews indicated that a more lenient visitation policy may speed recovery and reduce patient resistance to treatment. Nurse interviews suggested that initiating psychological therapy more often within the inpatient program may help patients make the transition from hospital to home, reducing the high rate of multiple readmissions.

Recognizing (as this program did and as Hutton advocates) the importance of allowing patients to use personal space as a means of expression may be a starting point for therapeutic communication.17 Most of the adolescents said they would also appreciate it if nurses ate something with them, rather than simply supervise their meals-even if it was just a snack. They suggested that sharing mealtimes might make them seem more "casual." The exception was Megan, who thought it would not make the situation any more comfortable and might cause patients to compare their meals to that of the nurse. Nurses Donna and Mandy did, on occasion, share a meal with the patients. Although Nurse Gabrielle did not, she said she thought it might be a good idea.

Outcomes. As a result of this study, in July 2008, the program employed a master's prepared nurse with a graduate certificate in child and adolescent mental health to specifically represent this group of patients and educate and support other nurses providing care to them.

The challenge before us is to reform the culture of inpatient behavioral anorexia treatment programs without losing their benefits. The unpleasant facts that the prison metaphor exposes about the existing culture of such programs suggest goals for reform: promoting therapeutic relationships, reducing readmissions, and lowering the level of frustration felt by nurses and patients alike.

REFERENCES