Stats

Clinical Practice in Pediatric Psychology Early Weight Loss in Adolescent Weight Management: The Role of the Home Environment Katherine E. Darling, Manfred H. M. van Dulmen, Geoffrey E. Putt, and Amy F. Sato Online First Publication, February 3, 2022. http://dx.doi.org/10.1037/cpp0000434

CITATION Darling, K. E., van Dulmen, M. H. M., Putt, G. E., & Sato, A. F. (2022, February 3). Early Weight Loss in Adolescent Weight Management: The Role of the Home Environment. Clinical Practice in Pediatric Psychology. Advance online publication. http://dx.doi.org/10.1037/cpp0000434

Early Weight Loss in Adolescent Weight Management: The Role of the Home Environment

Katherine E. Darling1, Manfred H. M. van Dulmen1, Geoffrey E. Putt2, and Amy F. Sato1

1 Department of Psychological Sciences, Kent State University 2 Pediatric Psychiatry and Psychology, Akron Children’s Hospital, Akron, Ohio, United States

Objective: Successful weight loss early in treatment is a key factor for long-term weight management success in adolescence. Yet prior research has not examined factors in the home environment related to risk for increased weight status as potential predictors of early weight management success. The primary goal of the present study was to explore the impact of modifiable household factors on baseline weight status and early weight status change among adolescents participating in an outpatient weight management program to identify clinical targets of early intervention. Method: Parents of adolescents (N = 188) presenting to an interdisciplinary weight management clinic within a children’s hospital completed measures at initial presentation. Objective adolescent weight status was collected at baseline and 2-month follow-up (n = 97). Results: Household chaos was significantly associated with weight status at presentation to the clinic, F(3, 181) = 3.85, p = .011. Similarly, household chaos was the only unique predictor of weight change from baseline to 2 months, F(3, 92) = 3.03 p = .033. Conclusions: Household factors, particularly household chaos, have often been overlooked in the adolescent obesity literature but are likely key contributors to early intervention response in a clinical weight management. This study highlights the importance of assessing and intervening on chaos in the household as higher levels of chaos may negatively impact early treatment outcomes among adolescents with obesity.

Implications for Impact Statement The present study suggests that household chaos may be a key modifiable factor impeding adolescent weight loss early in treatment. Although not often considered in prior research, chaos may serve as a key target for future weight management interventions to promote improved outcomes during adolescence.

Keywords: obesity, adolescent, weight management, household chaos, home environment

Katherine E. Darling https://orcid.org/0000-0002- 1858-4644 Geoffrey E. Putt https://orcid.org/0000-0002-5567-

8762 Katherine E. Darling is now affiliated with the Department

of Psychiatry and Human Behavior, Alpert Medical School of Brown University. None of the authors have any conflicts of interest to report.

The research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Kent State University requires their students to submit an electronic copy of their master’s/doctoral thesis to their digital archival repository so that it will be openly available,

in full, to anyone, free of charge. Portions of this article are adapted from the dissertation of Katherine E. Darling that can be found at http://rave.ohiolink.edu/etdc/view?acc_num= kent1559729147579083 and here modified from the original and presented in peer-reviewed format for the first time. A version of this data was also presented as a virtual poster presentation at the 2020 Society for Pediatric Psychology Annual Conference.

Correspondence concerning this article should be addressed to Katherine E. Darling is now at the Weight Control and Diabetes Research Center, The Miriam Hospital, 196 Richmond Street, Providence, RI 02903, United States. Email: [email protected]

1

Clinical Practice in Pediatric Psychology

© 2022 American Psychological Association ISSN: 2169-4834 https://doi.org/10.1037/cpp0000434

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

Obesity is a major public health concern among adolescents in the United States, with approxi- mately 40% of adolescents currently classified as being overweight or obese (Skinner et al., 2018). Obesity during adolescence is associated with a number of health concerns and negative psychoso- cial outcomes, as well as increased morbidity and mortality into adulthood (Gurnani et al., 2015). De- spite promising treatments for pediatric obesity, many youth presenting to treatment are unsuccess- ful in changing their weight status (Wilfley et al., 2018). Specifically, recent data from the Pediatric Obesity Weight Evaluation Registry (POWER) identified an average change in percent of the 95th percentile of body mass index (BMI) of �2.0 for adolescents ages 12–14 years after 4–6 months of treatment (Kumar et al., 2019). A decrease of 5% indicates clinically meaningful change (Kumar et al., 2019). This finding suggests that most adoles- cents do not achieve clinically significant weight loss even after participating in a multicomponent weight management intervention. Data from POWER have also identified that early BMI reduc- tion is significantly associated with long-term weight management success across a wide age range (4–18 years) of children (Gross et al., 2019). Given these findings, it is particularly important to identify potential predictors of weight loss, includ- ing early weight loss, among adolescents present- ingforweightmanagement. Prior research has focused on patient factors as

predictors of weight loss success in adolescent weight management programs (Braet, 2006; Jela- lian et al., 2008). These individual factors include higher baseline BMI, older age (within childhood), and male gender all being associated with greater weight loss success. However, these individual- level factors are not modifiable and cannot be directly targeted through behavioral weight control (BWC) interventions. The American Academy of Pediatrics recommends involving parents in BWC through behavioral strategies such as limit setting, modifying the home environment, and reducing barriers (Bean et al., 2020; Spear et al., 2007). As children transition to adolescence, there is an increase in autonomy and desire for independence; however,thehouseholdenvironmentstillhasasub- stantial impact of health behaviors (Bean et al., 2020; Shrewsbury et al., 2011). Research to date has not examined specific facets of the home envi- ronment that may predict weight status and early treatment success of adolescents seeking BWC intervention.

Home Environment

One such aspect of the immediate home environ- ment is food insecurity, the experience of limited or uncertain access to food (Nord et al., 2007), which has been related to increased rates of being over- weight (20.8% compared to 15.6% in those without food insecurity) among adolescents ages 12–17 years(Caseyetal.,2006).Anumberofmechanisms have been proposed to explain this association, including increased availability of energy-dense foods (Drewnowski, 2004), a cycle of restriction during times of decreased food availability leading toovereatingwhenfoodismoreplentiful,andmeta- bolic changes (Casey et al., 2006; Drewnowski, 2004). Clinically, food insecurity may be a key intervention point, potentially with families receiv- ing education about purchasing low-cost healthy foods, to promote BWC effectiveness. However, studiesexaminingtheassociationbetweenpediatric obesityandfoodinsecurityinadolescencehavetyp- ically not been conducted within the context of a clinical sample. Instead, these studies have focused on convenience samples and national surveys (Lar- son & Story, 2011). Therefore, this study sought to examine whetherfoodinsecurity maybe one aspect of the home environment that predicts early BWC treatmentsuccessforadolescents. Time constraints—a lack of time to maintain

healthyeatingandexercisebehaviorsduetoworkor other commitments—have been associated with increased adolescent weight status and lack of healthy eating in the family (Hearst et al., 2012; Jabs & Devine, 2006). Specifically, time scarcity has been associated with less healthful diet choices, including increased consumption of convenience or ready-prepared foods and fast food, and a decrease in family meals and food preparation at home (Jabs & Devine, 2006). BWC requires a significant time investment. Families are typically asked to attend regularappointments,makespecificandmeasurable goals, and perform actions toward those goals (O’Connor et al., 2017). Given the added time- related demands associated with implementation of BWCintervention components(e.g., timeforphysi- cal activity, time for grocery shopping and cooking, time for attending appointments), families with increased time-related barriers may be less likely to besuccessfulinearlyBWC. Householdchaos,orthephysicalandsocialdisor-

der at home, is an additional factor likely related to early BWC success. Chaos within the household is indicated by running late, lack of planning, and a

2 DARLING, VAN DULMEN, PUTT AND SATO

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

noisy home. A chaotic home environment has been related to risky health outcomes for adolescents, including increased risk for smoking, drinking, and substance use (Chatterjee et al., 2015). A recent overview of the literature has identified the impor- tance of behavioral and social routines (opposite of chaos)inthefamilytotreatingobesitythroughchild- hood and adolescence (Hart et al., 2020). In young children, household routines have been related to weight status in both cross-sectional and interven- tion research (Anderson & Whitaker, 2010; Haines et al., 2013). Specifically, in a cross-sectional analy- sis, exposure to higher levels of household routines (i.e., family meals, adequate sleep, and limiting screen time) was associated with lower prevalence of obesity as compared to young children not exposed to these routines (Anderson & Whitaker, 2010).Althoughthiswasanationallyrepresentative sample taken from a cohort study, it was limited by the use of single-item measurements for each rou- tine.Inaddition,thiscohortstudywaslimitedbythe inclusionofonlychildrenaged4years(Anderson& Whitaker, 2010). In children from kindergarten to eighth grade, fewer family routines have been related to increased probability of childhood obesity (Anderson, 2012). A randomized study of young children (2–5 years) found that an intervention to promote household routines led to decreased BMI (Hainesetal.,2013).Researchindicatingtheimpor- tanceofanorganizedhouseholdforpositiveweight- related outcomes among younger children has not yet been examined within adolescents, especially earlyinweightmanagementtreatment.

The Current Study

The present study extends the literature on pre- dictorsofadolescentweightstatusandearlyweight change in the context of a hospital-based pediatric BWC program by examining the impact of three modifiable household factors. The first aim was to examinethe associationbetweenhouseholdfactors (food insecurity, time constraints, and household chaos) and weight status at initial presentation (i.e., baseline). It was hypothesized that higher levels of food insecurity, time constraints, and household chaos would be associated with higher weight sta- tus at baseline. The second aim was to examine the impact of these household variables on weight sta- tus change over the first 2 months of the interven- tion. It was hypothesized that higher levels of food insecurity,timeconstraints,andhouseholdchaosat baseline would all be associated with less weight

statuschangeoverthefirst2monthsofweightman- agementintervention.

Method

Participants and Procedures

Participants included 205 adolescents (10–18 years) presenting to a tertiary care pediatric BWC program in the midwestern United States and their parents/guardians.Parentsandadolescentswereel- igible to participate if both the parent and adoles- cent (a) spoke English fluently and (b) did not have an identified learning disorder or cognitive disabil- ity preventing them from completing question- naires(perparentreport).

Procedures

Aspartofstandardclinicalcare,eachnewfamily was mailed a packet with psychosocial measures andbriefinformationaskingthefamilytocomplete the measures prior to their first clinic appointment. If patients did not complete measures prior to their initial appointment, they were provided with an additional copy of measures and asked to complete them immediately prior to the first clinic appoint- ment. Consent/assent were obtained in a private area by trained research assistants at patients’ first clinic visit. Psychosocial measures were collected once, at the beginning of treatment, with anthropo- metric data collected at each appointment within the clinic. This study was approved by the Akron Children’sHospitalInstitutionalReviewBoard.

Intervention

Each patient was seen by a psychologist, exercise physiologist, dietician, and medical provider (physi- cian or nurse practitioner) at each 2-hr appointment within the clinic. During the intake evaluation, the focus was on assessing current activity level, eating patterns, and psychological factors related to their weight status. Brief psychoeducation and goal set- ting were conducted during each visit. Although no treatment manual was utilized, at each appointment, providers generally reviewed information previ- ously presented, discussed problems or barriers that familiesfacedrelatedtobehaviorchange,introduced new topics related to weight loss, and set new goals for the upcoming month. Intervention topics were chosen with input from both providers and patients, covering topics relevant to the adolescent while also

ROLE OF THE HOME ENVIRONMENT IN ADOLESCENT BWC 3

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

allowing for collaborative session planning. Exer- cise physiologists introduced components of physi- calactivityappropriatefortheadolescents’leveland provided strategies for increasing physical activity. Dieticians focused on planning a healthy, nutritious diet and providing feedback on food intake. Specific self-monitoring targets varied between participants (e.g., only tracking fruit and vegetable intake, track- ing all consumption, tracking physical activity) depending on treatment targets. Medical providers reviewed lab work, which was taken at the first appointment and when clinically indicated, and dis- cussedpotentialhealthcomplications.Psychologists collaborated with families to discuss specific goals and strategies for behavioral changes suggested by otherdisciplines(e.g.,behavioralreinforcement,set- ting specific reasonable goals) as well as review other behavioral strategies for weight loss (e.g., slowingpaceofeating,self-monitoring,contingency reinforcement). Families were asked to schedule appointments

approximatelyoncepermonthforatotalof6months, with extension as appropriate for each family. Find- ings from the present study focused on baseline mea- surement and measurement at the adolescent’s third appointment (approximately 2 months out from the initial intake). This time point was selected to be con- sistent with prior research demonstrating treatment success at 2 months as a predictor of longitudinal treatmentsuccess(Unicketal.,2015).

Measures

Demographics

Parents completed a demographic question- naire concerning characteristics of the parent and adolescent participating in the clinic. This ques- tionnaire included items regarding both parental and adolescent age, sex, race/ethnicity, educa- tional attainment, as well as parental employment status, occupation, and marital status. Demo- graphic items (i.e., parent-report marital status, employment status, educational attainment, and occupation)wereusedtocalculate socioeconomic status (SES) using the Hollingshead four-factor index (Hollingshead, 1975). Education was coded on a 7-point scale, and occupation was rated on a 9-point scale, with a range of total scores from 8 to 66. All calculations were conducted following the guidelines set forth by Hollingshead (1975), with higherscoresreflectinghigherSES.

Adolescent Weight

Adolescent height and weight were measured objectively at each clinic appointment. Trained staff members collected measurements while patients were wearing light clothing and no shoes. Weight in kilograms was measured using a Scale- tronix digital scale, and height in centimeters was measured using a Seca stadiometer. Measurements were used to calculate BMI (kg/m2) at baseline and each subsequent time point. Prior literature has identified limitations of other commonly used met- ricsofadolescentweightstatus(e.g.,zBMI).There- fore,BMIandchangeinBMIoverthecourseofthe intervention were used to measure adolescent weight status, consistent with prior research in this area(Grossetal.,2019).

Food Insecurity

Parent-report household food insecurity was measured via the validated USDA Core Food Se- curity Module—Short Form (CFSM-SF; Bickel et al., 2000). This six-item scale measures level of food insecurity within the past year. The CFSM- SF was developed from the 18-item CFSM through identification of the six indicators that approximated the categories of the initial food se- curity measure (Bickel et al., 2000). Affirmative responsesaresummed,with higherscores indicat- inghigherfoodinsecurity.Specifically,scorescan range from 0 (food secure) to 10 (very low foodse- curity or food insecurity with hunger). An exam- ple item is “In the last 12 months, did you ever eat lessthanyoufeltyoushouldbecausetherewasnot enoughmoneytobuyfood?”Itemresponsesdiffer between items and include “yes” and “no” (three items); “often true,” “sometimes true,” and “never true” (two items); and “almost every month,” “some months but not every month,” and “only 1 or 2 months” (one item). The CFSM has been found to be valid and reliable in previous studies (Bickeletal.,2000).

Time Constraints

Timeconstraintswithinthefamilyweremeasured via parent report on the time constraints subscale of the Barriers to Pediatric Weight Management Scale (Darling et al., 2018). This five-item subscale assesses parents’ perceptions of their time con- straintsinrelationtotheirabilitytoimplementhealth family behaviors. Items were rated from 1 (strongly disagree) to 5 (strongly agree), and a mean of all

4 DARLING, VAN DULMEN, PUTT AND SATO

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

items was calculated for the overall time constraints scorewitharangefrom1,indicatinglowertimecon- straints, to 5, indicating higher time constraints. An example item is “It is hard for me to find the time to prepare healthy foods at home.” This measure has shown good convergent validity and appropriate in- ternal consistency within prior normative and clini- calsamples(Darlingetal.,2020,2018).

Household Chaos

The parent-reported Confusion, Hubbub, and Order Scale—Short Version (CHAOS) is a six-item measure used to identify chaos and disorder within the household (Hart et al., 2007; Matheny et al., 1995). Parents rated items on a 5-point scale from 1 (definitelyuntrue)to5(definitelytrue).Ameanscore of all items was calculated, with a range of scores from 1, indicating lower chaos, to 5, indicating higher chaos. Items on this scale capture general chaos (four items; e.g., “It’s a real zoo in our house” and “You cannot hear yourself think in our home”) and routines (two items; i.e., bedtime and screen time routines). The CHAOS scale has shown strong psychometric characteristics and has been shown to be valid and reliable for adolescents ages 10–18 years(Chatterjeeetal.,2015;Mathenyetal.,1995).

Data Analytic Plan

Missing data was handled using listwise dele- tion. All models include age, sex, and SES as cova- riates based on associations in prior literature. Sex was coded as 0 for men and 1 for women. Tests of normality assumptions (i.e., skew and kurtosis) wereconductedforallvariablesofinterest.Prelimi- nary analyses included independent-samples t tests to examine differences on demographic variables between adolescents presenting less than three ses- sions to those that attended at least three sessions. Correlationswereconductedbetweenallstudyvar- iables and continuous demographic variables (i.e., ageandSES). Hierarchical linear regression was used to test

the first aim of the study, the impact of household factors on adolescent weight status at presentation to treatment. Age, sex, and SES were included as covariates in Step 1 of the model, with all three household factors, food insecurity, household chaos, and time constraints, entered into Step 2 of the model to predict adolescent BMI. Similarly, hierarchical linear regression was used to examine the impact of the household variables on weight

change from baseline to Session 3. Paralleling Aim 1, age, sex, and SES were included as covariates in Step 1 of the model, with the three household fac- tors entered into Step 2 of the model to predict changeinadolescentBMI.

Results

Preliminary Analyses

Adolescents (M age = 13.26, SD = 2.22) who participated were primarily non-Hispanic White (70.7%), and over half were female (59%). All variables were normally distributed, and no transformations were required. Missing data ranged from 0% to 4.3%, with 16 adolescents excluded from the final sample due to missing data. Within the final sample, 188 adolescents attended the first appointment, 127 attended a second appointment (67.5%), and 97 attended a third appointment (51.6%). Further descriptive statistics for the sample are listed in Table1.Inde- pendent-samples t tests were conducted to examine differences between participants who attended less than three appointments (i.e., dropping out prior to the 2-month follow-up) to those that remained engaged in treatment at the third appointment. Nosignificant differences weredetected between groups. Of adolescents who attended all three appointments, 65.1% of the sample lost weight; however, the average BMI change (M = �.23, SD = 1.60) did not reach statistical significance (p = .21). Of the adolescents who attended all three appointments and lost weight, the average BMI change was �.91 (SD = 1.26; p , .001). Correlations between all variables can be found in Table2.

Primary Analyses

The hierarchical linear regression model testing Aim1examinedwhetherhouseholdvariablessignifi- cantly predicted baseline BMI, controlling for age, sex, and SES. Household factors significantly pre- dicted baseline BMI. Specifically, the second step of the regression (including all four household factors) was statistically significant, F(3, 181) = 3.85, p = .011, explaining 5.3% of the variance in adolescent BMI beyond age, sex, and SES. Household chaos wastheonlysignificantpredictorofweightstatus(p= .007).FullregressionresultsaredisplayedinTable3.

ROLE OF THE HOME ENVIRONMENT IN ADOLESCENT BWC 5

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

The hierarchical regression model testing Aim 2 foundthathouseholdfactorssignificantlypredicted BMI change beyond age, sex, and SES, F(3, 92) = 3.03, p = .033, explaining 8.9% of the variance in adolescent BMI change beyond age, sex, and SES. Consistent with findings for Aim 1, household chaos was the only significant predictor of BMI change(p=.017).PleaseseeTable3forfullregres- sionresults.

Discussion

Prior research has demonstrated that weight loss earlyinpediatricobesitytreatmentisthebestpredictor of treatment success using a national registry of weight management programs across the United States (Gross et al., 2019). Despite this work, prior research has not yet examined the factors that may promote initial treatment progress for adolescents

Table 2 Correlations and Descriptive Statistics for Study Variables

Variables 1 2 3 4 5 6 7 M SD

1. BMI at Session 1 — 35.90 6.78 2. BMI at Session 2 (N = 127) .98** — 36.66 6.50 3. BMI at Session 3 (N = 97) .97** .97** — 36.58 6.75 4. Adolescent age �.12 �.08 �.08 — 13.26 2.22 5. SES (Hollingshead four-factor index) �.15* �.17 �.15 .04 — 37.08 15.66 6. Family time constraints .03 �.03 .04 .19* .15* — 2.69 .82 7. Food insecurity (Core Food Security Module) .13 .12 .14 �.11 �.27** .00 — 1.05 1.78 8. Household chaos .23* .21* �.17 .19* .02 .38** �.05 3.42 .67 Note. N = 188 unless otherwise noted. BMI = body mass index; SES = socioeconomic status. *p , .05. **p , .01.

Table 1 Descriptive Characteristics for Adolescent and Parent Demographic Variables

Characteristics N (or M) % (or SD)

Adolescent characteristics Biological sex Female 111 57.1% Male 77 42.9%

Ethnicity Hispanic or Latino 5 2.7% Not Hispanic or Latino 169 89.9% Don’t know/declined to respond 14 7.4%

Race White/Caucasian 133 70.7% Black/African American 29 15.5% Other/don’t know 30 13.4%

Age M = 13.26 SD = 2.22 BMI at Session 1 M = 35.90 SD = 6.78 BMI at Session 2 (N = 127) M = 36.66 SD = 6.50 BMI at Session 3 (N = 97) M = 36.57 SD = 6.75

Parent characteristics Biological sex Female 163 86.7% Male 22 11.7% Declined to respond 3 1.6%

Education (of clinic-attending parent) Did not complete high school 8 4.1% High school graduate 57 29.3% Partial college (at least 1 year) or specialized training 59 27.7% Standard college or university graduation 49 25.0% Graduate professional training (graduate degree) 23 12.2%

Note. All percentages may not equate to 100 due to rounding. N = 188 unless otherwise noted. BMI = body mass index.

6 DARLING, VAN DULMEN, PUTT AND SATO

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

participating in a hospital-based BWC program. This is especially important to identify clinical targets for earlyinterventiontopromoteimprovedtreatmentsuc- cessoverthefirstfewsessions. Household chaos was the only significant pre-

dictor of both baseline weight status and early treatment success. Chaos within the household, characterized by high levels of disorganization, background stimulation, lack of family routines, and absence of predictability in the home, may impede sustained health-related behaviors. De- spite the demonstrated importance of household routines (e.g., mealtime, screen time, bedtime) on weight status in preschool- and school-age chil- dren (Anderson et al., 2017; Anderson & Whi- taker, 2010; Haines et al., 2013) the importance of routines and stability (which minimize chaos in the household) has been widely ignored as chil- dren transition into adolescence. Expanding upon prior work in adolescents relating a chaotic home environment to risky health outcomes (Chatterjee et al.,2015),thisstudy isthe first tolink household chaos to clinical BWC outcomes for adolescents. Further, the present study focuses on general chaos in the household, which may be a more accurate assessment of the broader family envi- ronmentascomparedtoaspecificmeasuresuchas householdroutines. Beyondinterventionsfocusedonhouseholdrou-

tines, a recent scoping review identified an absence of interventions with the primary aim of addressing

household chaos (Marsh et al., 2020). Clinicians and clinical researchersshould considerfocallytar- geting both general household chaos and routines within initial sessions of weight management inter- vention to increase the likelihood that families are able to engage in other weight management strat- egies (e.g., self-monitoring, stimulus control). For someyouthseekingweightmanagement,clinicians may consider working with families to identify the underlying causes of chaos in their home and de- velopfamily-specificstrategiestodecreasingchaos earlyinBWC. Contrary to hypotheses in the present study, nei-

ther food insecurity nor time constraints were related to either baseline weight status or weight change early in treatment. Although both of these household factors were included in the present study due to their identified associations with pedi- atric weight status (Casey et al., 2006; Hearst et al., 2012), these associations were not identified within the present study. Prior research concerning food insecurityandweightstatusinadolescentshasbeen mixed. Although some studies have identified an association (Casey et al., 2006), others have found that the prevalence of overweight/obesity are high in adolescents with food insecurity, despite no direct association between these variables (Eisen- mann et al., 2011). Further, although time con- straints have been associated with adolescent weight status (Hearst et al., 2012), this has not been explored in the context of BWC. Prior work

Table 3 Regression With SES and Psychosocial Variables Predicting Baseline Weight Status and Weight Change

Variables B SE b R2 DR2

BMI at intake Step 1: .110 .110** Age .88 .21 .29** Sex �1.52 .97 �.11 SES �.07 .03 �.26*

Step 2: .163 .053* Household chaos 2.04 .75 .20** Food insecurity .52 .27 .14 Time constraints �.59 .61 �.07

BMI change Step 1: .013 .013 Age �.06 .08 �.07 Sex .34 .35 .10 SES �.00 .01 �.03

Step 2: .102 .089* Household chaos .67 .28 .27* Food insecurity �.07 .10 �.08 Time constraints �.18 .23 �.09

Note. SE = standard error; BMI = body mass index; SES = socioeconomic status. *p , .05. **p , .01.

ROLE OF THE HOME ENVIRONMENT IN ADOLESCENT BWC 7

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

regarding both food insecurity and time constraints has been primarily conducted in nonclinical sam- ples and hasincluded youth acrossthe weight spec- trum (Casey et al., 2006; Hearst et al., 2012). It is possible that food insecurity and time constraints function differently for adolescents presenting to a clinical program as compared to prior nonclinical samples. Although autonomy is increasing for adoles-

cents, parents still maintain a key role in weight management success (Spear et al., 2007). For example, approximately 66% of food intake occurs in the home, even in adolescence (Poti & Popkin, 2011), and parents are continuing to be primarily responsible for the types of foods avail- able and eaten (Reicks et al., 2015). Understand- ing the role of parents in treatment and assessing the impact of specific parent variables on adoles- cent weight outcomes is vital (Bean et al., 2020). Future research should consider aspects of the home environment related to household chaos, such as family conflict, parenting behaviors, and family dysfunction, in early treatment success for adolescentsenrolledinBWC. Findings from the present study should be con-

sideredinlightofseverallimitations.First,allpsy- chosocial measures were assessed from the perspective of caregivers. Future research should employ multiinformant (i.e., both youth and care- giver report) and multimethod techniques. For example, coding objective food insecurity within the home may provide insight into both the objec- tive and subjective experiences of food insecurity (Webb et al., 2006). Additionally, the present study focused on the examination of early BWC outcomes among adolescents specifically given the paucity of research in this age group.Although analyses were conducted controlling for age, out- comeswerenotexaminedinthecontextofdevelop- mental stage (e.g., younger vs. older adolescents), and this study did not allow for comparison of school-age versus adolescent youth. Further, the sample primarily consisted of younger adolescents, and findings may differ for older adolescents with more autonomy. Future research should explore the rangeofdevelopmentaldifferences. This study was limited in the inclusion of only

adolescents and caregivers that could speak Eng- lish. Although thelocationin whichthis studywas conducted includes a relatively small proportion of individuals that do not speak English (a small midwestern city), this is not representative of the United States at large, and findings should be

explored in a more heterogeneous population. Similarly,adolescents in the present study primar- ilyidentifiedasWhite.Theproportionofindividu- als from different racial and ethnic backgrounds was representative of the broader community in which the study occurred; however, findings from the present study may not be generalizable to a more heterogenous sample. Future research should consider the impact of both language and race/ethnicity on household factors predicting early weight management success for adoles- cents, especially given that adolescents from racially and ethnically diverse outcomes are at higher risk for obesity (Skinner et al., 2018). Findings from this study should also be inter-

preted considering the high rate of attrition and lack of statistically significant BMI change limit- ingtheabilitytopredictchangeinadolescentBMI. As the goal of the present study was to examine earlysuccessinintervention,onlythreetimepoints representing approximately 2 months elapsed were included. However, almost half of patients had dropped out of the intervention by the third appointment. Unfortunately, this finding is similar toareviewofotherhospital-basedpediatricweight management programs, with previous studies reporting attrition rates greater than 50% (Sallinen Gaffka et al., 2013). Novel approaches to enhance treatment engagement, improve patient retention, and improve BWC outcomes should also be explored within hospital-based pediatric BWC clinics. Further, research should explore whether intervening on factors related to early weight man- agement success (i.e., household chaos) during the first treatment sessions may promote increased engagement and retention in weight management programs.

Conclusions

Overall, household chaos at presentation to BWC predicted baseline weight status and early treatment weight loss. These findings highlight the importance of addressing the impact of household chaos within pediatric BWC, especially for an ado- lescent population in which attention to the home environment is often neglected. Findings from the current study present multiple considerations for futureresearchonpredictorsofearlytreatmentsuc- cess for adolescents in order to capitalize on these and promote improved success in hospital-based pediatricBWCinterventions.

8 DARLING, VAN DULMEN, PUTT AND SATO

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

References

Anderson, P. M. (2012). Parental employment, fam- ily routines and childhood obesity. Economics and Human Biology, 10(4), 340–351. https://doi.org/10 .1016/j.ehb.2012.04.006

Anderson, S. E., Sacker, A., Whitaker, R. C., & Kelly, Y. (2017). Self-regulation and household routines at age three and obesity at age eleven: Longitudinal analysis of the U.K. Millennium Cohort Study. International Journal of Obesity, 41(10), 1459–1466. https://doi.org/10.1038/ijo .2017.94

Anderson, S. E., & Whitaker, R. C. (2010). House- hold routines and obesity in U.S. preschool-aged children. Pediatrics, 125(3), 420–428. https://doi .org/10.1542/peds.2009-0417

Bean, M. K., Caccavale, L. J., Adams, E. L., Burnette, C. B., LaRose, J. G., Raynor, H. A., Wickham, E. P., & Mazzeo, S. E. (2020). Parent involvement in ado- lescent obesity treatment: A systematic review. Pedia- trics, 146(3), Article e20193315. https://doi.org/10 .1542/peds.2019-3315

Bickel, G., Nord, M., Price, C., Hamilton, W., & Cook, J. (2000). Guide to measuring household food security. U.S. Department of Agriculture, Food and Nutrition Service, Office of Analysis Nutrition, and Evaluation.

Braet, C. (2006). Patient characteristics as predictors of weight loss after an obesity treatment for chil- dren. Obesity, 14(1), 148–155. https://doi.org/10 .1038/oby.2006.18

Casey, P. H., Simpson, P. M., Gossett, J. M., Bogle, M. L., Champagne, C. M., Connell, C., Harsha, D., McCabe-Sellers, B., Robbins, J. M., Stuff, J. E., & Weber, J. (2006). The association of child and household food insecurity with childhood over- weight status. Pediatrics, 118(5), e1406–e1413. https://doi.org/10.1542/peds.2006-0097

Chatterjee, A., Gillman, M. W., & Wong, M. D. (2015). Chaos, Hubbub, and Order Scale and health risk behaviors in adolescents in Los Angeles. The Journal of Pediatrics, 167(6), 1415–1421. https://doi.org/10.1016/j.jpeds.2015 .08.043

Darling, K. E., Ruzicka, E. B., Fahrenkamp, A. J., Putt, G. E., & Sato, A. F. (2020). Parent-perceived barriers to child weight management: Measure val- idation in a clinical sample. Families, Systems, & Health, 38(2), 130–138. https://doi.org/10.1037/ fsh0000480

Darling, K. E., Sato, A. F., van Dulmen, M., Flessner, C., & Putt, G. (2018). Development of a measure to assess parent perceptions of barriers to child weight management. Childhood Obesity, 14(2), 89–98. https://doi.org/10.1089/chi.2017.0171

Drewnowski, A. (2004). Obesity and the food envi- ronment: Dietary energy density and diet costs.

American Journal of Preventive Medicine, 27(3), 154–162. https://doi.org/10.1016/j.amepre.2004.06 .011

Eisenmann, J. C., Gundersen, C., Lohman, B. J., Garasky, S., & Stewart, S. D. (2011). Is food inse- curity related to overweight and obesity in children and adolescents? A summary of studies, 1995- 2009. Obesity Reviews, 12(5), e73–e83. https://doi .org/10.1111/j.1467-789X.2010.00820.x

Gross, A. C., Kaizer, A. M., Kelly, A. S., Rudser, K. D., Ryder, J. R., Borzutzky, C. R., Santos, M., Tucker, J. M., Yee, J. K., & Fox, C. K., & the POWER Work Group. (2019). Long and short of it: Early response predicts longer-term outcomes in pediatric weight management. Obesity, 27(2), 272–279. https://doi.org/10.1002/oby.22367

Gurnani, M., Birken, C., & Hamilton, J. (2015). Childhood obesity: Causes, consequences, and management. Pediatric Clinics, 62(4), 821–840. https://doi.org/10.1016/j.pcl.2015.04.001

Haines, J., McDonald, J., O’Brien, A., Sherry, B., Bottino, C. J., Schmidt, M. E., & Taveras, E. M. (2013). Healthy habits, happy homes: Randomized trial to improve household routines for obesity pre- vention among preschool-aged children. JAMA Pediatrics, 167(11), 1072–1079. https://doi.org/10 .1001/jamapediatrics.2013.2356

Hart, C. N., Jelalian, E., & Raynor, H. A. (2020). Be- havioral and social routines and biological rhythms in prevention and treatment of pediatric obesity. American Psychologist, 75(2), 152–162. https://doi.org/10.1037/amp0000599

Hart, S. A., Petrill, S. A., Deckard, K. D., & Thompson, L. A. (2007). SES and CHAOS as envi- ronmental mediators of cognitive ability: A longitu- dinal genetic analysis. Intelligence, 35(3), 233–242. https://doi.org/10.1016/j.intell.2006.08.004

Hearst, M. O., Sevcik, S., Fulkerson, J. A., Pasch, K. E., Harnack, L. J., & Lytle, L. A. (2012). Stressed out and overcommitted! The relationships between time demands and family rules and parents’ and their child’s weight status. Health Education & Behavior, 39(4), 446–454. https://doi .org/10.1177/1090198111426453

Hollingshead, A. B. (1975). Four-factor index of social status. Yale University, New Haven. (Unpub- lished manuscript).

Jabs, J., & Devine, C. M. (2006). Time scarcity and food choices: An overview. Appetite, 47(2), 196–204. https://doi.org/10.1016/j.appet.2006.02.014

Jelalian, E., Hart, C. N., Mehlenbeck, R. S., Lloyd- Richardson, E. E., Kaplan, J. D., Flynn-O'Brien, K. T., & Wing, R. R. (2008). Predictors of attrition and weight loss in an adolescent weight control program. Obesity, 16(6), 1318–1323. https://doi .org/10.1038/oby.2008.51

Kumar, S., King, E. C., Christison, A. L., Kelly, A. S., Ariza, A. J., Borzutzky, C., Cuda, S., &

ROLE OF THE HOME ENVIRONMENT IN ADOLESCENT BWC 9

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.

Kirk, S. (2019). Health outcomes of youth in clini- cal pediatric weight management programs in POWER. The Journal of Pediatrics, 208(01), 57–65. https://doi.org/10.1016/j.jpeds.2018.12.049

Larson, N. I., & Story, M. T. (2011). Food insecurity and weight status among U.S. children and fami- lies: A review of the literature. American Journal of Preventive Medicine, 40(2), 166–173. https:// doi.org/10.1016/j.amepre.2010.10.028

Marsh, S., Dobson, R., & Maddison, R. (2020). The relationship between household chaos and child, parent, and family outcomes: A systematic scoping review. BMC Public Health, 20(1), Article 513. https://doi.org/10.1186/s12889-020-08587-8

Matheny, A. P., Jr., Wachs, T. D., Ludwig, J. L., & Phillips, K. (1995). Bringing order out of chaos: Psychometric characteristics of the confusion, hubbub, and order scale. Journal of Applied Devel- opmental Psychology, 16(3), 429–444. https://doi .org/10.1016/0193-3973(95)90028-4

Nord, M., Andrews, M., & Carlson, S. (2007). ERS report summary: Household food security in the United States, 2006 (Report No. ERR-49). Eco- nomic Research Service, U.S. Department of Agriculture.

O’Connor, E. A., Evans, C. V., Burda, B. U., Walsh, E. S., Eder, M., & Lozano, P. (2017). Screening for obesity and intervention for weight manage- ment in children and adolescents: Evidence report and systematic review for the U.S. Preventive Services Task Force. JAMA: Journal of the Ameri- can Medical Association, 317(23), 2427–2444. https://doi.org/10.1001/jama.2017.0332

Poti, J. M., & Popkin, B. M. (2011). Trends in energy intake among U.S. children by eating location and food source, 1977-2006. Journal of the American Dietetic Association, 111(8), 1156–1164. https:// doi.org/10.1016/j.jada.2011.05.007

Reicks, M., Banna, J., Cluskey, M., Gunther, C., Hongu, N., Richards, R., Topham, G., & Wong, S. S. (2015). Influence of parenting practices on eating behaviors of early adolescents during inde- pendent eating occasions: Implications for obesity prevention. Nutrients, 7(10), 8783–8801. https:// doi.org/10.3390/nu7105431

Sallinen Gaffka, B. J., Frank, M., Hampl, S., Santos, M., & Rhodes, E. T. (2013). Parents and pediatric weight management attrition: Experiences and rec- ommendations. Childhood Obesity, 9(5), 409–417. https://doi.org/10.1089/chi.2013.0069

Shrewsbury, V. A., Steinbeck, K. S., Torvaldsen, S., & Baur, L. A. (2011). The role of parents in pre- adolescent and adolescent overweight and obesity treatment: A systematic review of clinical recom- mendations. Obesity Reviews, 12(10), 759–769. https://doi.org/10.1111/j.1467-789X.2011.00882.x

Skinner, A. C., Ravanbakht, S. N., Skelton, J. A., Perrin, E. M., & Armstrong, S. C. (2018). Preva- lence of obesity and severe obesity in U.S. children, 1999–2016. Pediatrics, 141(3), Article e20173459. https://doi.org/10.1542/peds.2017-3459

Spear, B. A., Barlow, S. E., Ervin, C., Ludwig, D. S., Saelens, B. E., Schetzina, K. E., & Taveras, E. M. (2007). Recommendations for treatment of child and adolescent overweight and obesity. Pediatrics, 120(Suppl. 4), S254–S288. https://doi.org/10.1542/ peds.2007-2329F

Unick, J. L., Neiberg, R. H., Hogan, P. E., Cheskin, L. J., Dutton, G. R., Jeffery, R., Nelson, J. A., Pi- Sunyer, X., West, D. S., & Wing, R. R., & the Look AHEAD Research Group. (2015). Weight change in the first 2 months of a lifestyle intervention predicts weight changes 8 years later. Obesity, 23(7), 1353–1356. https://doi.org/10.1002/oby.21112

Webb, P., Coates, J., Frongillo, E. A., Rogers, B. L., Swindale, A., & Bilinsky, P. (2006). Measuring household food insecurity: Why it’s so important and yet so difficult to do. The Journal of Nutrition, 136(5), 1404S–1408S. https://doi.org/10.1093/jn/ 136.5.1404S

Wilfley, D. E., Hayes, J. F., Balantekin, K. N., Van Buren, D. J., & Epstein, L. H. (2018). Behavioral interventions for obesity in children and adults: Evidence base, novel approaches, and translation into practice. American Psychologist, 73(8), 981–993. https://doi.org/10.1037/amp0000293

ReceivedMarch10,2021 RevisionreceivedNovember16,2021

AcceptedDecember10,2021 n

10 DARLING, VAN DULMEN, PUTT AND SATO

T hi s do cu m en ti s co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly fo rt he

pe rs on al us e of th e in di vi du al us er an d is no tt o be

di ss em

in at ed

br oa dl y.