Family-based treatment (FBT) is the leading evidence-based treatment for adolescents with eating disorders [1]. This manualized treatment consists of three phases: in the first phase, parents are empowered to take charge of all aspects of their ill child’s eating until, in the second phase, once the eating disorder symptoms and thoughts have begun to subside, the child gradually takes back responsibility for his or her eating. In the third phase, the family and treatment team ensure that the patient is back on track with typical adolescent development [2]. Phase 1 is perhaps the most challenging phase, as resistance from the eating disorder can be quite strong at the outset of treatment. As such, the majority of the treatment manual focuses on the first phase of FBT, with a particular emphasis on the first two sessions, which can be challenging to implement and are essential to ensuring that treatment commences successfully.

Whereas Phase 1 is fairly structured, Phase 2 is less so. The goals of weight restoration and parental control over eating are quite clear in Phase 1. In Phase 2, the gradual handing back of control over eating to the adolescent can proceed in any number of ways, depending on the age of the adolescent and the family structures, routines, and preferences. For example, some families may enter Phase 2 by monitoring a meal (often school lunches) less frequently, whereas others may proceed by offering the adolescent limited choices between equally caloric options for a snack. The focus in Phase 2 should remain squarely on issues related to the eating disorder, and the treatment manual outlines the goals and objectives of this phase [2]. However, in the 250 + page manual, only about 30 pages are dedicated to Phase 2, while over 140 pages address Phase 1. Although there is a sample Phase 2 session included in the manual, the variability of Phase 2 could lend itself to additional examples being included in future versions of the manual. As Phase 2 is less prescriptive than Phase 1, FBT therapists sometimes feel that they have less of a “road map” to guide them in implementing the treatment approach as they reach Phase 2. This leaves some room for the possibility that FBT therapists will “fill in the gaps” with their own approaches when faced with clinical situations that are either not covered in the manual, or are more open to clinical interpretation given the increased flexibility inherent in Phase 2.

A study of uptake of FBT conducted by Couturier et al. [3] found that the majority of therapists assessed did not conduct FBT per the manual, citing reasons such as not wanting to weigh the patient, feeling anxious about conducting the family meal in session 2, and being uncomfortable with providing nutritional guidance without a dietitian. However, when attempting to implement evidence-based treatments, even well-intentioned deviations from the manual mean that the clinician is no longer practicing the form of treatment that has garnered research evidence to support its use. This may be particularly problematic when key components of the treatment approach are not being practiced with fidelity. In FBT, the core tenets of parental empowerment, parents being united against the eating disorder, not criticizing the patient, and externalization of the illness predicted weight gain [4]. However, another study of FBT found that treatment adherence was not associated with percent expected body weight at end of treatment [5], although adherence decreased over Phases 2 and 3 of treatment. Similarly, a study assessing level of agreement on FBT treatment fidelity among expert raters, peer raters, therapists, and parents, found high levels of agreement among raters for Session 1, but found that levels of agreement decreased as treatment progressed [6]. The authors suggest that this may be because FBT becomes less prescriptive as times goes on. Likewise, a dissemination study of FBT found that treatment fidelity was rated as 5 or higher on a 7-point scale 72% of the time in Phase 1, but only 47% of the time in Phase 2 [7]. There is evidence outside the FBT literature that fidelity to a cognitive-behavioral treatment model and adherence to an evidence-based protocol are associated with improved treatment outcomes [8], suggesting that therapist fidelity is an important variable to assess, and Phase 2 thus far has received comparatively little attention in the FBT literature.

The purpose of the current exploratory study was to assess Phase 2 practices of clinicians providing FBT and to compare these practices to the guidance offered in the manual to determine to what extent clinicians adhere to or drift from the primary Phase 2 interventions of FBT. The Family Therapy Fidelity and Adherence Check (FBT-FACT) [9] has been used in previous studies [5] to assess fidelity to FBT, but was designed to assess Phase 1. Thus, a survey specifically assessing Phase 2 interventions was developed for the current study.

Method

Procedure and participants

A Google Forms survey was sent to members of three groups: the Academy for Eating Disorders Family-Based Treatment Special Interest Group listserv, the Maudsley FBT Therapists Facebook Group (a private Facebook group with 101 members), and a nationwide group of mental health providers who participate in a monthly FBT consultation meeting, consisting of 25 members. The survey began by telling therapists: “We are looking for information around what clinicians actually focus on/implement during Phase 2”. The survey asked whether participants were providing FBT in an outpatient setting and included three open-ended questions about readiness for Phase 2, readiness for Phase 3, and common challenges encountered during Phase 2. Fourteen items then listed various possible topics addressed during Phase 2 and asked participants how often they included these topics as a focus of their Phase 2 sessions. Items were scored from 0 (0% or never) to 4 (100% of the time). Items were chosen by SJ and LM based on the identification of targets of independent eating, clinical issues that frequently arise during Phase 2, and issues deemed to be important for practicing clinicians during this phase [10]. Demographic information for the therapists was not collected. Institutional Review Board approval was not required for the current study.

Results

Twenty-seven therapists responded to the survey. Two reported that they did not provide FBT in an outpatient setting. One reported not currently providing outpatient FBT but had in the past. The remaining providers were currently providing FBT in an outpatient setting. Table 1 provides an overview of all topics and the number and percentage of therapists endorsing each category. The topics that were endorsed as being covered 100% of the time in Phase 2 by at least 50% of respondents included: encouraging age-appropriate independent eating (66.7%), nutrition and health maintenance (55.6%), encouraging the patient to engage in non-physical activities and interests outside of the eating disorder (55.6%), and overall increased flexibility (51.9%).

Table 1 Family-based treatment phase 2 survey responses
Full size table

Two topics listed in the survey are issues that are commonly addressed in dialectical behavior therapy (DBT) but are not described in the FBT manual: addressing patient interpersonal skill deficits, and addressing patient distress tolerance and emotion regulation skill deficits. However, these were endorsed as being addressed at least some of the time by 81.5% and 92.6% of clinicians, respectively. There is also a focus in Phase 2 of maintaining the progress made in Phase 1. “Nutrition and health maintenance” was endorsed as being covered 100% of the time in Phase 2 by 55.6% of therapists. Other important areas of focus in Phase 2—encouraging the patient to engage in non-physical activities and interests outside of the eating disorder, and overall increased flexibility—were highly endorsed, but no items central to Phase 2 were reported by 100% of therapists as being covered 100% of the time in Phase 2. Return to physical activity was reported by all therapists as being addressed to some extent in Phase 2, but was endorsed by only five therapists as being addressed 100% of the time. This may have depended in part on how therapists interpreted this item. An increase in physical activity during Phase 2 is typical of most patients, but not all will return to organized physical activity, such as participation in sports.

When asked about Phase 2 challenges, the pacing of Phase 2 was the most common response. Eleven clinicians raised the issue of families moving too quickly or too slowly through Phase 2: e.g., “going at the right pace—not too fast and not too slow”, “introducing too much too fast”. Six therapists mentioned parental burnout or anxiety: e.g., “parental anxiety and difficulty tolerating less than 100% seamless transition”, “sometimes parents are extremely nervous about allowing patients to make any choices with food”. Six therapists mentioned struggles with finding the right level of independence for the adolescent: e.g., “knowing how much to monitor/how much independence to give”, “patients often want more independence than they are ready for, given their ED symptoms”. Only two therapists mentioned concerns about comorbid diagnoses that might interfere with progress, and none mentioned deficits in interpersonal skills, emotion regulation, or patient distress tolerance, which is notable given the number of therapists who endorsed addressing these issues in Phase 2.

Discussion

The purpose of this exploratory study was to assess the frequency with which various topics are covered in Phase 2 by therapists, the majority of whom were currently practicing FBT. The primary focus of Phase 2 in FBT is the gradual returning of control over eating to the adolescent. This was assessed in the current study with the item “encouraging age-appropriate independent eating”. Although this was the item that was endorsed by the greatest number of therapists as being covered 100% of the time in Phase 2, and no therapists reported covering this 0% of the time, nine therapists reported addressing this between 25 and 75% of the time in treatment. Although unfortunately there was not an opportunity for therapists to elaborate on their answers, it is curious that one-third of respondents reported not addressing this topic 100% of the time in their work, as it is at the core of Phase 2 of FBT [2] and a necessary step in helping the adolescent return to a normal, healthy life. Indeed, if a patient is not ready to return to age-appropriate independent eating, then a family cannot progress through Phase 2. Identifying social barriers to independent eating is part of returning control over eating back to the adolescent, but was endorsed by fewer than half of therapists as being covered 100% of the time in Phase 2.

Externalizing the eating disorder is a key tenet of FBT and is introduced in the first session. Externalization refers to the separation of the illness from the ill child, emphasizing to the family that the child has been overtaken by a powerful disorder, which is impacting his or her thoughts, feelings, and behaviors when it comes to issues of food, eating, shape, and weight, and that the ill child is not being purposely difficult when he or she does not eat. Parental burnout also emerges as a common topic in FBT, particularly in Phase 1. Although both topics are emphasized in Phase 1, both can be addressed as needed throughout treatment, and externalization of the eating disorder continues as a theme throughout FBT. Most therapists reported addressing externalization at some point in Phase 2, and all therapists reported addressing parental burnout to some extent in Phase 2.

Several survey items addressed topics that are not included in the FBT manual, including topics that are normally addressed in DBT [11], such as patient deficits in interpersonal skills, distress tolerance, and emotion regulation. Although these items were endorsed less frequently than the items reflecting the core objectives of Phase 2 according to the FBT manual [2], the majority of therapists reported addressing these issues at some point in treatment. It is possible that addressing these topics is helpful in Phase 2, but it is not part of the FBT manual and reflects a certain amount of therapist drift from the evidence-based manualized treatment. FBT has a very behavioral focus on food and eating issues, and it is possible that some therapists grow fatigued with this approach and want to “mix-up” treatment by incorporating elements of other modalities. However, the laserlike focus on food and eating issues may be one of the reasons that FBT is so effective. Future research should explore this.

Likewise, FBT is not designed to address body image concerns. Anecdotally, these can improve over the course of treatment, although one study found that shape and weight concerns do not improve over the course of FBT [12], and a meta-analysis of treatments for anorexia nervosa found that specialized treatments were not more effective than comparator treatments in bringing about psychological change [13]. In the FBT manual, body image is only mentioned in Phase 3, primarily in the context of reviewing typical adolescent developmental concerns with the family [2]. However, the majority of therapists reported addressing body image issues at some point during their Phase 2 work. Providing psychoeducation around adolescent development and assessing the presence of other issues/disorders are typically done in Phase 3 [2], but most therapists reported addressing these issues to some extent in Phase 2. Relapse prevention is not explicitly stated as a goal of any phase in the FBT manual, although anecdotally it is often addressed in Phase 3 of treatment. However, most therapists reported addressing relapse prevention in Phase 2.

A study of cognitive processing therapy for posttraumatic stress disorder (PTSD) found that greater numbers of fidelity-consistent modifications were associated with greater reductions in PTSD and depressive symptoms [14]. The authors suggested that appropriate adaptations that did not interfere with the key components of cognitive processing therapy could improve outcomes, whereas fidelity-inconsistent modifications, such as “integration of other treatment elements can detract from adherence to the elements of the protocol intended to address these key elements” (p. 365). It is possible that incorporating other treatment elements, such as DBT techniques, detracts from FBT’s focus on behavioral change, whereas focusing on relapse prevention, for example, may not be in conflict with key FBT principles. Future research should expand on our limited knowledge of treatment adherence and treatment outcome, particularly in FBT, as only one study has assessed this [5], and should assess not only treatment adherence but therapist competence, as this has been shown to influence outcome in other disorders [15]. Additional avenues for future research involve more closely assessing the reasons behind apparent deviations from the manual. For example, some therapists may feel the need to address body image concerns as a way of supporting patients with identified social barriers that are interfering with a return to independent eating. Thus, the deviation could be utilized in the spirit of accomplishing Phase 2 FBT goals. This could be investigated further via interviews with FBT therapists. Future research should also further assess dissemination of FBT. Efficacious evidence-based treatments are useful to the extent that they can be successfully disseminated outside of the centers that developed them. Research has found that FBT can be disseminated into private practice settings [16] and results in significant decreases in hospital readmission rates when utilized after inpatient treatment [17]. Additional studies investigating dissemination of FBT and improved patient outcomes are needed. Finally, given the frequency with which therapists in the current study endorsed using DBT techniques, incorporation of DBT into FBT should be further assessed. Several studies have examined this [18, 19] but have not directly compared the combination of FBT and DBT to either approach alone.

Overall, the therapists’ answers seem to reflect general adherence to the FBT manual, but with some notable exceptions (e.g., addressing interpersonal skills deficits) that reflect therapist drift and incorporation of therapeutic techniques from other treatment modalities. Given the variability with which Phase 2 can be implemented, the FBT manual understandably does not address every avenue taken by families. An in-depth analysis of Phase 2 interventions that are consistent with FBT principles may be helpful for practicing clinicians and may enhance the utility of the current manual as well as increase treatment fidelity.

Strengths and Limitations.

Limitations of the current study include lack of information on therapist response rate and lack of information on demographic characteristics of responding therapists. Importantly, a process exists to certify providers in FBT, but information on whether or not the therapists participating in the current study were certified in FBT was not collected. It is possible that certified providers are more likely to follow FBT per the manual than those who are not. Information was also not available on how and where the therapists practiced, how long they had been practicing, and how they learned FBT. In addition, given that only three participants reported not currently providing outpatient FBT, chi-square analyses examining differences in patterns of endorsement according to current practice of FBT were not possible. The survey used in the study is an additional limitation, as it was created for the purposes of this study and has not been validated. It is possible that therapists utilize additional treatment modalities that were not assessed in the survey, thus, the conclusions that can be drawn from this study are somewhat limited. Additional limitations include the small sample size, low response rate, the descriptive nature of the analyses, and the cross-sectional design. Finally, we could not link therapist practices to treatment outcome. This is a significant limitation that should be addressed in future studies. Strengths of the study include the first analysis of mental health providers’ practices in Phase 2 of FBT, contributing to our understanding of where and how they deviate from the manual.

What is already known on this subject?

Therapist drift during Phase 1 of FBT is common, possibly influencing the effectiveness of the treatment. This drift is common despite the FBT manual providing clear guidelines as to the implementation of Phase 1. However, there is less guidance in the manual regarding Phase 2, and no information available on how therapists implement this phase.

What this study adds?

This study is the first to assess therapist practices in Phase 2 of FBT. Core components of Phase 2 were not always addressed 100% of the time, and elements of other treatment modalities were often incorporated. Future studies should determine to what extent therapist drift impacts treatment outcome.