Adapting Interpersonal Psychotherapy for the Prevention of Excessive Weight Gain in Rural African American Girls

Abstract

Objective To obtain focus group data regarding the perspectives of rural African American (AA) girls, parents/guardians, and community leaders on obesity, loss of control (LOC) eating, relationships, and interpersonal psychotherapy for the prevention of excessive weight gain (IPT-WG). Methods 7 focus groups (N = 50 participants) were moderated and the transcripts analyzed by Westat researchers using widely accepted methods of qualitative and thematic analysis. A session was held with experts in health disparities to elucidate themes. Results Participants understood LOC eating; however, they had culturally specific perceptions including usage of alternative terms. Relationships were highly valued, specifically those between mothers and daughters. IPT-WG program components generally resonated with participants, although modifications were recommended to respect parental roles. Experts interpreted focus group themes and discussed potential barriers and solutions to recruitment and participation. Conclusion Findings suggest that adapting IPT-WG may be acceptable to rural AA families. This research is the first step in developing a sustainable excessive weight gain and binge eating disorder prevention program for rural AA adolescents.

Forty percent of African American (AA) youth are overweight (body mass index, BMI, kg/m², 85–95th percentile for age and sex) or obese (BMI ≥ 95th percentile for age and sex (CDC, 2000); Ogden, Carroll, Kit, & Flegal, 2012), and rates range from 26 to 42% among low-income rural-dwelling AA children (Liu et al., 2012). Because excess weight in youth predicts adult obesity (Field, Cook, & Gillman, 2005) and rural AAs have limited access to healthcare and are at increased risk for poor health outcomes (U.S. Department of Health and Human Services [DHHS] Health Resources and Services Administration [HRSA]), excessive weight gain prevention in this population is critical. Yet, preventative efforts among AA youth have been largely unsuccessful (Black et al., 2010). Targeted prevention programs that focus on modifiable risk factors for excessive weight gain may hold promise, particularly in AA youth (Austin et al., 2011). Although previous qualitative studies have examined perspectives of AAs on obesity, physical activity, and nutrition, reports rarely consider psychological risk factors for excess weight gain, such as loss of control (LOC) eating, and their associated mechanisms, including interpersonal relationships and negative affect, that may inform specific prevention programs. Elucidation of such risk factors from the community’s perspective is important to inform culturally relevant prevention.

LOC eating is a developmental risk factor for excess weight gain (Tanofsky-Kraff et al., 2009) and partial/full-syndrome binge eating disorder (BED; Tanofsky-Kraff et al., 2011). Unlike binge eating (APA, 2000), LOC eating refers to the experience of a lack of control over eating, regardless of the amount of food consumed (Tanofsky-Kraff, 2008). While BED is less common in youth, LOC eating is reported in up to 45% of overweight adolescent girls (Tanofsky-Kraff, 2008). The prevalence of LOC is comparable (Austin et al., 2008) or greater (Swanson, Crow, Le Grange, Swendsen, & Merikangas, 2011) in AAs compared with other racial/ethnic youth. LOC is also endorsed with substantial prevalence in low-income urban and rural communities (Field, Colditz, & Peterson, 1997). Targeting LOC eating in AAs may serve as one selective approach to reducing excessive weight gain in this population (Tanofsky-Kraff, 2012).

Interpersonal psychotherapy for the prevention of excessive weight gain (IPT-WG) was developed as a targeted approach to preventing excessive weight gain in adolescent girls who endorse BED symptoms, such as LOC eating (Tanofsky-Kraff, 2007b). IPT-WG is a brief time-limited therapy based on the theory that interpersonal difficulties lead to negative moods that trigger LOC eating, as individuals use food to cope with negative emotions (Tanofsky-Kraff et al., 2007b). IPT-WG maintains the key components of traditional IPT (Klerman, Weissman, Rounsaville, & Chevron, 1984; Weissman, Markowitz, & Klerman, 2000) and extends on IPT-Adolescent Skills Training for the prevention of depression (Young, Mufson, & Davies, 2006) and IPT for BED (Wilfley, MacKenzie, Welch, Ayres, & Weissman, 2000). IPT-WG focuses on developing strategies for improving interpersonal functioning by identifying links between interpersonal issues (e.g., role disputes, grief), negative affect, and LOC eating (Wilfley et al., 2000) through psycho-education, skill building (Klerman et al., 1984; Weissman et al., 2000), communication analysis, and role playing (Young et al., 2006).

There are a number of reasons that IPT-WG may be especially appropriate for AAs who endorse LOC eating. First, IPT-WG emphasizes positive relationships and familial connectivity, which are highly valued among AAs (Hill, 1999). Secondly, because AAs may be at increased risk for stressful life events, changes in family structure, and interpersonal distrust (Wilfley, Pike, & Striegel-Moore, 1997), the IPT-WG model may more aptly describe the triggers for disordered eating in this group (Wilfley et al., 1997). Indeed, IPT-WG focuses on helping individuals handle stressful events, including role transitions that result from changes in the family milieu and improving interpersonal trust and communication (Tanofsky-Kraff et al., 2007b). Unlike cognitive behavior therapy, IPT-WG does not focus on internalization of the thin ideal, which is rarely reported by AAs (Kelly, Bulik, & Mazzeo, 2011). Third, IPT-WG is well poised to address culturally specific challenges to LOC eating (e.g., large portions of ethnic foods) because the program targets the negative emotions associated with LOC eating, regardless of the particular situation.

Preliminary evidence suggests that IPT is a flexible therapy that may be adaptable for use in rural or remote settings (Bass et al., 2006; Bolton et al., 2003, 2007; Verdeli et al., 2003) and acceptable to diverse groups (Chui, Safer, Bryson, Agras, & Wilson, 2007; Mufson, Dorta, Moreau, & Weissman, 2004; Tanofsky-Kraff et al., 2010; Young et al., 2006). Although data in AA adolescents are limited, in a pilot study of >50% AA girls, those in IPT-WG reduced LOC eating episodes (p = .04, partial η² = .12) and experienced less-than-expected BMI growth at 1-year follow-up (B = 2.4, Wald Statistic = 4.8, (Exp)B = 10.5, p = .05) compared with those in health education (Tanofsky-Kraff et al., 2010). IPT targeting binge eating in overweight adults tends to result in weight maintenance or modest weight loss (Wilfley et al., 1993, 2002; Wilson, Wilfley, Agras, & Bryson, 2010). Taken together, if culturally adapted, IPT-WG may be an especially suitable therapy for rural AA adolescents with LOC eating.

Prevention efforts must be community based and tailored to specific racial/ethnic minority communities to improve acceptability, effectiveness, and sustainability (Kumanyika, 2010; Kumanyika et al., 2007). Cultural adaptations for AA children have been carried out in prior trials, such as The Hip Hop to Health Jr. (Fitzgibbon et al., 2005; Kumanyika, 2010). In this trial, during the formative phase, several elements for improving cultural appropriateness were identified and subsequently incorporated into the larger trial, including adjusting content to reflect environmental influences and incorporating a family component. Although cultural adaptations do not ensure significant effects, efforts that elicit community insights to inform program development may promote sustainability (Kumanyika et al., 2007). Community-based Participatory Research (CBPR) (Agency for Healthcare Research and Quality [AHRQ], 2002) is one method for improving sustainability (Kumanyika et al., 2007). CBPR is a collaborative approach involving community members and researchers in an effort to engage local communities in the design and dissemination of research products (AHRQ, 2002). Through this partnership, community members become invested in the dissemination and use of research findings, with the ultimate goal of sustainable change to reduce health disparities.

We, therefore, used CBPR methods to conduct focus groups with rural AA adolescent girls, parents/guardians, and community leaders to obtain detailed information on their perspectives regarding obesity, LOC eating, relationships, and the IPT-WG program as the first step in adapting IPT-WG for this underserved group. We hypothesized that participants would recognize relevant eating and weight constructs and subscribe to the IPT-WG model and program, and that culturally specific perceptions would be revealed.

Methods

Participants

The study was approved by the Institutional Review Boards at the Uniformed Services University of the Health Sciences and the University of Maryland Eastern Shore. A rural community sample of AAs primarily from Princess Anne and Salisbury communities in Maryland was recruited to participate in a focus group to discuss “thoughts, attitudes, and experiences related to body weight, eating, and communication.” In Princess Anne, MD, population density is 3,300; median household income is $32,200; and percentage of individuals below the poverty line is 33%. In Salisbury, MD, estimates are 30,300; $40,600; and 25% (respectively; U.S. Census Bureau, 2010). The communities were identified as rural due to limited access to healthcare, barriers to transportation, and geographic isolation (Eastern Shore Area Health Education Center; Maryland Department of Health and Mental Hygiene, 2007; U.S. DHHS HRSA). The study was restricted to AA adolescent females, as LOC eating is more prevalent in adolescents and girls (Tanofsky-Kraff, 2008). The threshold for LOC eating in the current study was set based on prior research, which demonstrates that even infrequent reports of LOC eating is predictive of excess weight (Stice, Cameron, Killen, Hayward, & Taylor, 1999) and body fat gain (Tanofsky-Kraff et al., 2006) and the onset of BED (Tanofsky-Kraff et al., 2011) in pediatric samples.

Procedure

Girls and parents/guardians were recruited through direct mailings, flyer postings, camp announcements, and local radio advertisements directed toward parents/guardians concerned about their daughter’s excessive weight gain and eating. Local community leaders were recruited by referral from site collaborators. Eligible girls (12–17 years) were AA, and overweight (BMI ≥ 85th percentile for age and sex) (CDC, 2000). LOC eating was assessed by telephone screen using one question (i.e., “Has there ever been a time when you were eating and you could not control what or how much you were eating?”) or the endorsement of features associated with LOC eating (Tanofsky-Kraff et al., 2007a). Features included (1) eating following negative emotions and/or a triggering event; (2) eating despite a lack of hunger; (3) eating initiated by or during a snack; (4) eating alone; (5) eating in secret; (6) dissociating while eating; or (7) feelings of guilt or shame after eating (see Table I). Girls who endorsed a lifetime occurrence of ≥1 LOC eating episode and/or ≥2 correlates of LOC were invited to participate with or without a participating parent/guardian. Adolescents provided written assent, and parents/guardians provided written consent for their daughters and themselves if they chose to participate. Community leaders, who also provided written consent, were eligible if they ever worked in a healthcare-related field and were individuals who would be likely to deliver the IPT-WG program in the community.

Table I.

Adolescents’ Endorsements of Loss of Control Eating, Episodes, and Correlates by Telephone Screen and Questionnaire

N = 21	Telephone screen	Eating Disorder Examination Questionnaire
LOC eating (%)	44	38
LOC eating episode days (mean, SD)	—	0.81, 1.5
LOC correlates (%)
Eating following negative emotions and/or a triggering event	37	—
Eating despite a lack of hunger	52	—
Eating that occurred during or initiated by a snack	74	—
Eating alone	30	—
Feelings of secrecy with regard to eating	41	—
A sense of “numbing” out while eating	15	—
Feelings of negative emotions following eating, such as guilt or shame	41	—

Note. Loss of control correlates were derived from previous work (Tanofsky-Kraff et al., 2007a). LOC = loss of control.

Eligible adolescents and parents/guardians participated in separate groups conducted concurrently. Girls were divided into age-appropriate groups (12–14 years and 15–17 years). Girls’ heights were measured in triplicate to the nearest 0.1 cm by the Detecto height scale (Cardinal Scale Manufacturing, Co., Webb City, MO). Weight (kg) and body fat (%) were measured to the nearest 10th by the Tanita BF 350 Body Composition Analyzer (Tanita Corporation of America, Inc., Arlington Heights, IL). BMI and BMI z-scores were subsequently calculated (CDC, 2000). Two cohorts of parent and child groups were completed in October and December 2011. The community leader group occurred in December 2011. Participants received $50, refreshments, and childcare if needed. Groups were held at the main campus of the University of Maryland Eastern Shore. Each session lasted ∼90 min. Following the focus groups, girls completed a questionnaire.

Measures

The Eating Disorder Examination Questionnaire (EDE-Q) (Fairburn & Beglin, 1994) is based on the semi-structured Eating Disorder Examination (Fairburn & Cooper, 1993), which is known as the gold standard for assessing disordered eating behavior, including LOC eating. Seven questions on the EDE-Q specifically queried LOC eating. The EDE-Q has been widely used in adolescent samples (Binford, Le Grange, & Jellar, 2005; Carter, Stewart, & Fairburn, 2001; Decaluwe, Braet, & Fairburn, 2003). There are limited reliability data on the EDE-Q in AA youth. However, in one sample of AA women, the EDE-Q demonstrated good internal consistency (Cronbach’s αs = .79–.91) (Kelly, Cotter, & Mazzeo, 2012). In a racial/ethnic minority community sample of adolescents and adults, the EDE-Q also demonstrated good internal consistency (Cronbach’s αs = .74–.86) (Pelaez-Fernandez, Javier Labrador, & Raich, 2012).

Focus Groups

To ensure rigor and limit researcher biases, moderators from Westat (Westat, Rockville, MD) were contracted to moderate focus groups. Westat is an international data collection agency that specializes in qualitative procedures. One moderator and one note-taker were assigned to each group. Moderators were matched on race/ethnicity, and experience ranged from 5 to 20 years. Guides were designed by the study team and refined in consultation with a senior analyst at Westat.

Adolescents

Adolescents were queried regarding their general eating behaviors, including healthy and unhealthy eating patterns. Participants were asked to discuss LOC and use of alternative terms, the frequency and context of LOC eating, and associated emotions. Adolescents also discussed their important relationships, relationship conflicts, and whether they recognized links between their relationships, mood, and LOC eating. They were queried regarding their opinions of the program and provided feedback on the skills and role-plays. Adolescents were asked to discuss incentives to participation.

Parents

Parents were queried regarding adolescents’ general eating behaviors, including healthy and unhealthy eating patterns. Participants were asked to discuss LOC and use of alternative terms, whether they observed LOC eating in their daughters, and associated emotions. Parents also discussed their daughters’ important relationships and relationship conflicts, including the parent–daughter relationship. Participants discussed whether they recognized links between relationships, mood, and LOC eating in their daughters. To assess the ways in which body shape ideals may impact perceptions of obesity and eating behaviors, parents were asked to describe their beauty ideals, how ideals have changed, and how body weight impacts their daughters. They were also queried regarding their definition of overweight/obesity. Parents were asked to discuss their opinions about the program, areas for improvement, barriers to participating, and the ideal qualifications of an IPT-WG leader.

Community Leaders

Leaders were queried regarding the general eating behaviors (including LOC eating) observed among adolescents in their community. Participants were asked to discuss LOC and use of alternative terms, whether they observed LOC eating in the community, and associated triggers. Leaders were asked to describe the community’s beauty ideals, how ideals have changed, and how ideals impact adolescents. They were also queried regarding their definition of overweight/obesity and associated views in the community. Community leaders were asked to discuss their opinions about the program, areas for improvement, the community’s potential response, and the ideal qualifications of an IPT-WG leader.

Analysis

Westat analyzed all data in several stages. Each focus group was audio recorded and transcribed verbatim. After each session, moderators and note-takers prepared transcripts by summarizing initially salient themes from each group. A lead analyst conducted a debriefing with the moderators and note-takers to compare, contrast, and discuss themes. An analyst used the initial themes identified by moderators and note-takers as a preliminary guide for textual examination of the transcript data. Using commonly accepted procedures for qualitative data analysis (Berkowitz, 1997; Bernard & Ryan, 1998; Glaser & Strauss, 1967; Miles & Huberman, 1994), senior analysts identified, refined, and verified the recurrent themes and patterns within (e.g., parent) and across (e.g., parent vs. teenager) groups for each area of interest: (1) general adolescent eating patterns; (2) LOC and binge eating; (3) relationships and the perceived links to LOC eating; and (4) reactions to the IPT-WG program. Inter-rater reliability was assessed according to qualitative research recommendations (Berkowitz, 1997; Golafshani, 2003; Morse, Barrett, Mayan, Olson, & Spiers, 2002). Two experienced Westat qualitative analysts independently read all transcripts several times, identifying relevant phrases, concepts, and themes that regularly or systematically emerged within and across the groups. Analysts independently coded data and then met to discuss and reconcile respective coding schemes. They did not use a software program, as use has not been shown to improve quality or rigor (Barry, 1998; Leech & Onwuegbuzie, 2007; Vander Putten & Nolen, 2008).

Expert Session

Four AA professionals with expertise in AA research and public health issues assisted researchers in understanding focus group themes and assessing reliability. Individuals were recruited from the USU Center for Health Disparities Community Research Outreach Workers Network (CROWN), a group of 20 experts in AA culture, mental health, and/or public health. Experts held doctorate degrees (n = 3) in clinical and/or child psychology or a master’s degree (n = 1) in social work. Experience with AAs ranged from 7 to 20 years. Study team members provided Experts with the transcript summary and themes from each focus group, in addition to six to eight questions to facilitate discussion. Experts reviewed the materials prior to participating in a 90-min teleconference to discuss impressions. The teleconference included Experts and study team members. The teleconference was audio-recorded, and a research team member summarized the session. Results were used to create the final summary of program recommendations.

Results

Demographics

Data were collected from 21 adolescents (M = 14.24 years, SD = 1.55 years), 19 mothers, 1 grandmother, 1 father, and 8 community leaders (N = 50). The study included four groups of adolescents, two groups of parent/guardians, and one group of community leaders. All adolescents were overweight or obese with a mean (SD) BMI z-score of 2.26 (0.47) and a mean (SD) BMI percentile of 97.20 (2.41). Body fat ranged from 33.40 to 56.10% (M = 46.00%, SD = 6.91%). Based on the EDE-Q, LOC eating episodes (days in the past month) ranged from 0 to 6, with a mean (SD) of 0.81 (1.47) (see Table I). Community leaders worked in various health-related professions such as child/family development, health education, nursing, phlebotomy, rehabilitation, and medicine. The highest education levels obtained by community leaders were associate’s (n = 1), master’s (n = 5), and doctoral (n = 2).

Focus Groups

Focus group themes were categorized into five areas (when applicable): eating patterns, LOC eating, body shape ideals/body weight, relationships, and general reactions to IPT-WG (see Table II for comparison of themes across groups). Westat reported that participants (1) appeared enthusiastic about the idea of the prevention program; (2) generally understood the links between mood and eating, even if they did not endorse it themselves; and (3) believed that they and/or others who were interested in an excessive weight gain prevention program would attend the program if cultural adaptations were made.

Table II.

Comparison of Themes Across Focus Groups

Themes	Adolescents	Parents	Community leaders
General eating patterns
Unhealthy eating patterns	✓	✓	✓
Mindless eating	✓	✓
Eating in secret		✓	✓
Binge/LOC eating
Various terms for LOC eating	✓	✓	✓
Triggers of LOC eating	Boredom, sadness, anger, and being upset with family members	Boredom, enjoyment of food, grief/loss	Strong negative and positive emotions, learned from parents
Body shape ideals/Body weight
Preference for larger body sizes	—	✓	✓
Generational change in body standards	—	Current preoccupation with weight and being thin	—
Distinction between “overweight” and “obese”	—	✓	✓
Relationships
Difficulties in relationships	“Drama” at school, conflicts with mother	Adolescent laziness	—
Desires for improvements in family relationships	✓	✓	—
IPT-WG
Reactions to IPT-WG program	Skill: “Use ‘I feel’ statements” and Skill: “Put yourself in other’s shoes” may not work	Reacted positively to the list of skills and the idea of the role-playing	Programs may be too rudimentary to engage teens
Desire for parent component	✓	✓	✓
Desire for behavioral component	✓	✓	✓
Factors influencing participation	Group number and incentive	Time and day of the week	—
Qualifications of an appropriate IPT-WG leader	—	Female, African American, young, patient, understanding, positive, nonjudgmental, someone who has lost weight successfully herself, and someone from outside of the immediate community	Cultural sensitivity and appropriateness, be able to interact with teenagers, formal education credentials of at least a Master’s degree, and weight and past experience with eating/weight control issues

Note. Groups that share a check (‘✓’) reported similar themes. Groups with a dash (‘—’) were not queried on the specific theme. Themes that differed across groups are provided in the relevant cell. Groups with no data on the specific theme are left blank.

LOC = loss of control; IPT-WG = interpersonal psychotherapy for the prevention of excessive weight gain.

Adolescent Themes

General Eating Patterns

Adolescents reported consuming unhealthy foods and sugary beverages. They reported skipping meals before and during school, and then overeating after school. They also reported skipping evening meals. Girls reported overeating during weekend activities (e.g., going to the mall) and described mindlessly eating while using the computer or watching television.

LOC Eating

Most adolescents were aware of the terms “binge” and “LOC eating,” but described the behavior as “being greedy,” “pigging out,” or “throwin’ down.” In the first set of focus groups, all girls endorsed LOC. In the second set, some endorsed LOC eating, while others did not. They associated LOC eating with feelings of anger, sadness, being upset with family members, boredom, idleness as well as positive emotions, like happiness. Younger adolescents reported experiencing feelings such as “regret,” “tired,” and “guilty” after overeating. Older adolescents also mentioned dissociating during overeating episodes, explaining that while eating, “it’s going right over your head, but after you’re done, you feel it.” Younger adolescents had more difficulty recognizing links between emotions and LOC eating behaviors, although both groups recognized the connection in others.

Relationships

Adolescents, especially older teens, recognized the links among LOC eating, relationships, and mood. Some girls reported difficulties in their relationships with schoolmates, younger siblings, and mothers, although most still regarded themselves as close to their mothers and saw these difficulties as a normal part of growing and establishing their own identities. They desired more quality time and improvements in their communication with parents and other family members.

Reactions to IPT-WG Program

Older adolescents understood the theoretical basis of the program and supported the concept that targeting relationships to improve emotions may impact eating behaviors for those prone to emotional and LOC eating, stating that “teens have mood swings everyday” and those moods may affect communication and consumption. Despite general enthusiasm for an IPT-based prevention program, there was concern that some program components and skills would not be applicable in the current form. For example, some girls disagreed with the skill encouraging the use of “I feel” statements, stating, “parents just don’t care how you feel. We would end up on punishment if we say something they don’t like.” Adolescents desired a parent component for the program and expressed an interest in role-playing communication if a parent were involved, explaining that parents would see “how hard they really are on us.” Adolescents also mentioned it would be helpful to role play how to “talk to friends.” Girls reported that a limited number of participants in the groups (∼5 members) and issuing some form of compensation (e.g., gift certificates) would be important factors in considering participation. Youth recommended that a behavioral component regarding healthy food choices and/or physical activity be incorporated.

Parent Themes

Adolescent Eating Patterns

Parents recognized that LOC eating is prevalent among adolescents in their community and described LOC as being “greedy.” Although all parents did not initially recognize a connection between eating behaviors, relationship issues, and emotions, parent reported links between LOC eating and feelings of grief after the death of a parent, relative, or close friend. Parents also attributed their daughter’s weight gain to “mindless” eating of unhealthy food, secretive eating, boredom, and daughters simply enjoying eating.

Body Shape Ideals/Body Weight

Parents noted their daughter’s physical appearance greatly impacted their lives, explaining that they are often excluded from social activities because of their weight. One parent noted that her daughter stays home to avoid participating in events where girls are wearing “tiny little clothes, short skirts and all.” Parents understood, but rejected, the pressure of mainstream body standards that emphasize thinness. They preferred terms such as “voluptuous” and “curvaceous” rather than “big,” “heavy,” or “fat.” They differentiated between overweight and obesity, with the former being related to one’s BMI and the latter signifying a heavy person with related medical problems. Although concerned about the health consequences, many were reluctant to classify their daughters as overweight or obese because they felt doing so would erode their daughters’ positive body- and self-images. Parents desired support in discussing eating and weight issues with daughters.

Relationships

Overall, parents felt positively about their relationships with their daughters. Conflicts involved “typical” adolescent issues (e.g., chores, homework). They desired to spend more quality time with them and improve communication—as one parent stated, “I am too critical sometimes.” They also desired better ways to discuss issues related to loss of a loved one.

Reactions to the IPT-WG Program

In general, parents were enthusiastic about a prevention program to target both excess weight gain and exacerbated disordered eating. Parents initially reacted positively to skills and role-playing; however, they believed that some program components should include a behavioral element related to nutrition and physical activity. They also desired a parent support group specifically devoted to discussing their daughters’ progress and any issues that might arise as a result of using new communication skills at home. According to the parents, important qualifications of IPT-WG leaders were being female, AA, relatively young (20s to mid 30s), patient, understanding, positive, nonjudgmental, someone who had lost weight successfully herself, and someone from outside of the immediate community because “you got to be able to trust that person not taking it back on the street about what’s going on.”

Community Leader Themes

Community Eating Patterns

Leaders explained that healthy eating is not a tradition in the community. Consumption of large amounts of unhealthy ethnic foods is viewed as part of the community’s cultural identity. Therefore, a reduction or elimination of these food choices may be viewed as “denying where I came from” or being “uppity.” Leaders highlighted components of the rural culture that may impact eating behaviors, such as poverty, physical and cultural isolation, and lacking a sense of self-efficacy. They reported observing adolescents skipping meals, but “gorg[ing]” when returning home from school. Community leaders believed the prevalence of LOC eating to be high among adolescents and the entire community. Leaders described LOC as “greedy” and eating without being hungry. Leaders attributed the eating behavior to “filling an emotional hole” and positive emotions. Several leaders also noted that LOC eating might be a “learned behavior,” as teens witness their parents reacting to emotions by losing control over their eating.

Community Body Shape Ideals/Body Weight

Leaders believed that the community standards of body shape differed from mainstream culture such that individuals prefer larger body sizes, which is often viewed as an indication of good health. One community leader noted that a “juicy” baby is regarded as healthy and grows into an adolescent “that has these overdeveloped breasts and hips [and] we say, ah, she’s fine. She’s grown into a real woman … [when] she’s really overweight.” “Overweight” was perceived to relate to BMI whereas “obesity” signified an overweight person with medical problems and physical limitations. Leaders reported that individuals in the community perceived body weight and size as largely genetic; therefore, overweight and obesity are often viewed as inevitable outcomes. One leader noted, “if the mother is that type that’s got the body shape that’s kind of thick, then the kids think that’s how they [will] grow up to be.” Therefore, community leaders noted that the recognition that overweight and obesity is a community health problem is low.

Reactions to the IPT-WG Model and Program

Community leaders understood the connection between negative emotions and exacerbated disordered eating in adolescents. They also believed that positive emotions, poverty, and lack of alternative physical activities impact eating patterns and excess weight gain. Leaders were skeptical about IPT-WG in its current form, explaining that it may be too rudimentary to engage teens. One community leader stated, “just bringing a group of teenagers from this community. I think they would get bored very quickly.” Therefore, they believed the program should include a behavioral component with information about nutrition and physical activity. Leaders emphasized the importance of parents being “cued in” to program components, so use of skills would not be viewed as violating cultural standards regarding the proper role of parents and children. They believed that IPT-WG leaders should be individuals who are culturally sensitive, able to interact with teenagers, formally educated with at least a Master’s degree, and who experienced eating/weight control issues. They noted that incentives or prizes would be important.

Experts Session Responses

Emotions and LOC Eating

Experts confirmed that the connection between negative emotions and LOC eating resonated with adolescents and parents. According to experts, although adolescents associated particular emotions (e.g., anger, boredom) with overeating and LOC, it was unclear whether adolescents recognized these emotions as precursors to LOC eating. Hence, they believed that when adapting IPT-WG, explicating the nature and rationale of the model might require additional time. The various terms used to describe LOC eating led experts to believe that adolescents likely viewed disordered eating as a choice or lack of will power rather than a behavior that may warrant intervention. To help girls understand LOC eating and feel comfortable expressing their inability to control their eating, experts emphasized the importance of building rapport and trust among leaders and group members.

Participation and Recruitment

Experts raised barriers that could hinder participation in a prevention program, including shyness, stigmatization, reluctance to share information, and failure to perceive eating or weight issues as problematic. They believed that linking eating and weight with known medical issues, such as high blood pressure, would be essential. They also suggested including psycho-education regarding the connection between emotions and LOC eating during recruitment. Although participants used alternative terms to describe LOC (e.g., “greedy”), experts discouraged the use of these terms in recruitment materials, as they could be stigmatizing. They felt that using alternative terms for LOC eating within IPT-WG sessions might elucidate the LOC construct.

IPT-WG Program

Experts concurred that suitable leaders should be individuals outside of the community as to avoid discomfort with sharing private information. They believed it would be critical for skills to be appropriately adjusted and communicated effectively by leaders. Experts explained that leaders should clearly describe skills and how to implement them within the culture. For example, the IPT-WG skill to use “I feel” statements to express thoughts or feelings is a universally effective interpersonal communication tool. Yet, the skill could be perceived as too assertive and, therefore, disrespectful towards parents within this population. Thus, specific terminology would require adjustment to reflect typical communication within the AA family. Experts believed that leaders should highlight eating precipitated by grief for IPT-WG participants when relevant. Likewise, discussions about body image should be related to girls by acknowledging the ways in which AA beauty ideals differ from the thin ideal of mainstream culture.

Experts believed that a parent component would be necessary to promote learning and use of the IPT skills and to facilitate adolescents’ progress through the program. They noted that a healthy eating (e.g., planning meals) and exercise (e.g., ways to be active) behavioral component would be beneficial as well. To incentivize the participants, experts suggested including excursions, such as a visit to a restaurant, which would embrace the cultural values of collectivism and celebration.

Discussion

The purpose of this study was to obtain qualitative data regarding the perspectives of overweight rural AA girls with LOC eating behaviors, their parents/guardians, and community leaders on obesity, LOC eating, relationships, and IPT-WG. Findings suggest that the program generally resonated with adolescents and parents; however, community leaders were less enthusiastic about IPT-WG in its current form. All results will be used to adapt IPT-WG to be more culturally appropriate. Notably, parents and community leaders rejected typical thin ideals and preferred larger body sizes, specifically as it related to their daughters. This finding is consistent with prior literature (Kronenfeld, Reba-Harrelson, Von Holle, Reyes, & Bulik, 2010). The need to uphold cultural values and maintain adolescents’ self-esteem will be retained in the adapted IPT-WG program. Additionally, parents and community leaders only regarded high body weights as problematic when accompanied by medical consequences. This finding is consistent with prior qualitative reports (Burnet et al., 2008) and trials in adults (Sbrocco et al., 2005) that suggest that AAs are motivated to lose weight primarily when health co-morbidities are present (Fallon et al., 2005). To effectively recruit AAs, it may be important to consider ways to elucidate the relationships among the IPT theoretical constructs and medical comorbidities to improve the program’s face validity.

A pervasive theme throughout the focus groups and expert session was the participants’ understanding of LOC eating. Participants often described LOC eating by using terms that imply control over the behavior. In adapting IPT-WG, it should be emphasized that LOC is not a “choice” behavior, but rather the result of a complex interaction of biological, psychological, and environmental factors (Wilfley, Vannucci, & White, 2010).

IPT-WG generally resonated with adolescents and parents if cultural adaptations were made. This finding supports prior data suggesting that IPT may resonate well with AAs (Chui et al., 2007). Although community leaders recognized the links between mood and eating, they were less enthusiastic about IPT-WG without the inclusion of a behavioral component. Girls and parents also unanimously recommended a behavioral component to teach skills for healthy eating and exercising. Prior research on behavioral modification for weight loss in AAs is inconsistent (Baranowski et al., 2003), and data in adults suggest that behavioral treatments may be less effective at reducing binge episodes than specialized treatment for BED (Wilson et al., 2010). Yet, building social support networks for healthy eating and physical activity across home, school, and community contexts may increase engagement in such beneficial behaviors (Salvy, de la Haye, Bowker, & Hermans, 2012) and sustain long-term weight loss (Wilfley et al., 2007). In addition, all participants advocated for a parent component. This adaptation may bolster IPT-WG, as parental involvement in family-based programs has been predictive of long-term success (Heinberg et al., 2010; Wadden, Butryn, & Byrne, 2004). IPT-WG could be adapted to include weight-related goals focused on eliciting social support for healthy behavior change and to involve parent facilitation and assistance in the practice of interpersonal skills. Experts also suggested including excursions (e.g., restaurant visits), which may incentivize and engage adolescents.

The skill promoting use of “I feel” statements might require a modification so parental authority is respected. The adapted manual will also address parent and child relationship conflicts and weight-related issues adolescents may experience. Parents noted a relationship between their daughter’s weight status and functional impairment in social domains. This finding is consistent with data demonstrating that obese AAs report a markedly reduced psychosocial quality of life as compared with Caucasian groups (Perez & Warren, 2012). IPT-WG, with its focus on improving mood and relationships, may be especially salient for addressing this issue in AAs. Further, rural participants may experience additional and unique social stressors that may impact mood and LOC compared with urban residents. IPT-WG is well poised to address various cultural and social contexts; the skills and role-plays will be adapted to reflect the components of the rural environment.

Study strengths include the use of qualitative analysis and the CBPR method. CBPR was critical in the initiation, recruitment, collection, and implementation of this research project. Indeed, community partners and participants were enthusiastic about the current project and future research. Additionally, we worked with expert focus group moderators and qualitative analysts (Westat). Study limitations include the small number of focus groups and limited generalizability, as rural AAs may not represent AAs in other environments. Further, because adolescents, on average, endorsed fewer than one LOC eating episode in the past month, data may not be representative of adolescents with more severe eating pathology. However, our goal was to examine adolescents with sub-threshold eating problems. Although the EDE-Q was used to assess disordered eating pathology, it has not yet been validated in AA youth. Due to the small sample size, reliability for the current study was not analyzed. To limit study burden for adults, who were not the central focus of group discussions or IPT-WG adaptations, we did not collect data on body weight or socioeconomic status. Characterization of the sample may be limited. Future research should also include a measure of health-related or physical quality of life, as it may be particularly salient in this population. In conclusion, IPT-WG appears to resonate with a rural AA community, but culturally appropriate adaptations are required. Future research steps include modifying the IPT-WG manual based on study data and conducting a pilot study to test the feasibility of the program in a rural AA community.

Funding

This work was supported by the National Institute of Minority Health and Health Disparities, P20 MD00505-02, Uniformed Services University of the Health Sciences Center for Health Disparities to TS and MTK.

Conflicts of interest: None declared.