Available online at www.sciencedirect.com

Nutrition Research 28 (2008) 13–20www.elsevier.com/locate/nutres

A nutrition and physical activity intervention promotes weight loss andenhances diet attitudes in low-income mothers of young childrenKristine C. Jordana, Jeanne H. Freeland-Gravesa,⁎, Deborah M. Klohe-Lehmana,

Guowen Caia, V. Saroja Vorugantia, J. Michael Proffitta, Henry J. Nussa,Tracey J. Milania, Thomas M. Bohmanb

aDivision of Nutritional Sciences, 1 University Station, A2700, The University of Texas at Austin, Austin, TX 78712bCenter for Social Work Research, The University of Texas at Austin, 1 University Station R5000, Austin, TX 78712

Received 16 June 2006; revised 16 November 2007; accepted 16 November 2007

Abstract

The purpose of this study was to evaluate a nutrition and physical activity program for reducing

⁎ Corresponding aE-mail address: jf

0271-5317/$ – see frodoi:10.1016/j.nutres.2

body weight and improving nutrition attitudes in mothers of young children. A convenience sampleof 114 intervention mothers and 33 comparison mothers was recruited from public health clinics andcommunity centers. Eligibility criteria included Hispanic, African American, or white ethnicity; bodymass index of at least 25 kg/m2; low income (b200% of the federal poverty index); and youngestchild aged 1 to 4 years. For intervention participants, height, weight, percentage of body fat, waistcircumference, demographics, nutrition attitudes, and dietary intake were measured at weeks 0 and 8;height, weight, percentage of body fat, and waist circumference were reassessed at 6 months.Overweight mothers in the comparison group provided anthropometric and demographic data atweeks 0 and 8. Changes in anthropometrics, attitudes, and dietary intake were evaluated inintervention mothers. Anthropometric data of intervention vs comparison group mothers wereexamined. Differences in anthropometrics and attitude scores between weight loss responders(≥2.27 kg) and nonresponders (b2.27 kg) were assessed at week 8. Intervention participants lostweight (x = −2.7 kg; P b .001), whereas comparison mothers gained a slight amount of weight(x= 0.1 kg) by week 8. Weight loss responders had healthier eating attitudes (5.6 vs 5.2; P b .01) andfewer perceived barriers (2.4 vs 2.9; P b .05) than nonresponders postintervention. In conclusion,this dietary and physical activity curriculum is a valuable resource for weight management programsserving low-income women.© 2008 Elsevier Inc. All rights reserved.

Keywords: Mothers; Obesity; Nutrition attitudes; Weight loss; Low-income

1. Introduction

The United States is facing an unprecedented epidemicof obesity. Approximately 22.9% of Americans wereconsidered obese, as defined by a body mass index (BMI)of 30 kg/m2 in 1988 to 1994 (NHANES); this increased to

uthor. Tel.:+1 512 471 0657; fax: +1 512 471 5844.g@mail.utexas.edu (J.H. Freeland-Graves).

nt matter © 2008 Elsevier Inc. All rights reserved.007.11.005

30.4% in 1999 to 2002 [1]. The percentage of overweightadults (BMI ≥25) in the United States has paralleled thistrend, rising from 55.9% to 65.1%. Overall, the prevalence ofobesity is higher among women [1], minorities [2], andpersons of lower socioeconomic status [3]. Among racialgroups, African American and Hispanic women incur higherrates of obesity (49.0% and 38.4%, respectively) than whites(30.7%) [1]. The concern with obesity is the increased riskfor a variety of chronic diseases, including diabetes mellitus,

14 K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

stroke, cancer, hypertension, sleep apnea, and cardiovasculardisease [4].

Most weight loss interventions have studied middle toupper income white women [5], but it is low-income womenwho comprise the majority group giving birth in the UnitedStates [6]. In addition, minorities may experience lower ratesof weight loss. For example, African American women, ascompared to whites, are less likely to participate in a weightloss program, have lower rates of weight loss in theseprograms, and are more likely to drop out [7]. Participationby low-income and minority women is problematic becauseof barriers such as program cost, lack of childcare, andfamily responsibilities [8].

Strategies for promoting weight loss in minorities haveincluded low-literacy materials [9,10], meal replacementshakes [7], culturally sensitive curriculums [11,12], church-based classes [13], improvements in nutrition knowledge[12,14], behavior modification [7,10], and aerobic exercise[15]. Components of behavior modification include self-monitoring with food diaries [12], social support [11], theuse of rewards [13], and self-efficacy training [16].

Nutrition attitudes may play a more important role indetermining food behavior than knowledge alone. Accordingto a meta-analysis of the literature by Axelson et al [17],there is a weak correlation between nutrition knowledge anddietary behavior (r = 0.10). Other factors, such as nutritionattitudes (beliefs about healthful foods, perceived barriers tohealthful eating, emotional cues associated with eating), tastepreferences, cultural norms, and behavioral capability, act atthe intrapersonal level to influence food choices [18].However, few studies have evaluated the associationbetween nutrition attitudes and weight loss in interventionstargeted at low-income populations. Fitzgibbon et al [19]conducted an obesity intervention with 24 African Americanmothers and daughters. The 6-week curriculum covered thefollowing topics: health consequences of obesity, fast food,fat content of foods, and nutrition labels. Althoughparticipants increased their nutrition knowledge by the endof the program, their nutrition attitudes did not improve.However, a companion article to this study found thatenhancements in healthful eating attitudes were associatedwith greater weight loss at week 8 in a triethnic sample(Hispanic, African American, white) of low-income mothers(n = 114) [20]. Other interventions have found improvementsin nutrition attitudes in the areas of dietary fat reduction[21,22], cancer prevention [23,24], and fruit and vegetableconsumption [25,26]. The justification to investigate therelationship between weight loss and nutrition attitudes issupported by recent research indicating that obese women at1 year postpartum reported greater barriers to healthfuleating and increased eating associated with emotional cuesthan normal weight women [27].

Clearly, effective interventions targeted toward low-income women are needed to combat the rising prevalenceof obesity and diabetes in the US. According to the SurgeonGeneral's call to action to prevent and decrease overweight

and obesity [28], the development of culturally appropriateinterventions to treat obesity is a high priority area for theUS public health system. Thus, the overall purpose of thisresearch is to assess the impact of the program on reducingbody weight in low-income mothers to decrease theprevalence of obesity in this population. In addition, ourstudy aims to identify (1) nutrition attitudes associated withweight loss and (2) correlations between nutrition attitudesand dietary factors.

2. Methods and materials

2.1. Design

The intervention was designed to provide pre- andpostmeasurements for a convenience sample of overweightor obese (BMI ≥25 kg/m2) mothers (n = 114) of youngchildren. In intervention subjects, height, weight, percentageof body fat, waist circumference, demographics, attitude, anddietary measures were obtained at weeks 0 and 8; programevaluation forms were collected at week 8; and height,weight, percentage of body fat, and waist circumferencewere reassessed at week 24 for a subsample of 93 participantswho were able to attend a follow-up measurement session.Participants were divided into responders (≥2.27 kg) ornonresponders (b2.27 kg) based on the amount of weightloss at week 8 [12]. A comparison group of mothers with asimilar BMI (≥25 kg/m2) volunteered to provide comparisondata for demographic and anthropometric measures at weeks0 and 8 only. Data on attitudes and dietary assessment wereunavailable for the comparison group. The comparison groupdid not receive any educational intervention. Weight loss atweek 8 was considered the primary end point; however,weight was reevaluated at week 24 in intervention motherswho could still be contacted to assess weight maintenance.All subjects gave informed consent before participation in thestudy, and the institutional review board at the University ofTexas (Austin, TX) approved the research protocol.

2.2. Subjects

Mothers of young children were recruited from SpecialSupplemental Nutrition Program for Women, Infants, andChildren (WIC) clinics, community centers, and churchesto participate in free weight loss classes or serve ascomparison subjects. Eligibility criteria for both groupsincluded African American, white, or Hispanic ethnicity;youngest child of 1 to 4 years; BMI of at least 25 kg/m2;low-income (qualification for WIC or food stamps orannual household income b200% of the federal povertyindex); and absence of breast-feeding (b5 min/d). The finalintervention sample size was 114 of 260 who came to thefirst class, for an attrition rate of 56%. For mothers whocompleted the 8-week program, the class attendance ratewas 87%. Dropouts did not differ from interventionsubjects on baseline characteristics, except that dropoutswere less likely to be living with a spouse/partner (61% vs

15K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

81%, P b .05) or to be employed (60% vs 74%, P ≤ .01).In addition, 33 overweight/obese comparison mothers ofsimilar demographic background provided anthropometricdata at weeks 0 and 8.

2.3. Intervention classes

Small group classes were held at clinics and communitycenters. Each 2-hour class incorporated recommendationsfor healthful eating, behavior modification, and physicalactivity that were based on suggestions following focusgroup discussions. The educational objectives for theintervention included the following: list the benefits ofweight loss, monitor portion sizes to allow for weight loss,self-monitor via food records and pedometers, interpretnutrition labels for energy and fat content, modify recipes todecrease calories, select nonfood rewards for attaining goals,promote improvement in attitudes about nutrition (increaseconcern for the nutritional content of foods, describe cuesassociated with emotional eating, propose solutions toaddress emotional eating and barriers to healthful eating,identify sources of stress that trigger eating, explain how tostop a food binge), demonstrate strength-building exercises,and describe techniques to help maintain weight loss.

The dietary recommendations were based on the USdietary guidelines that were designed to encourage healthfuleating habits and regular physical activity [29,30]. Thehealthful eating component consisted of menu plans withculturally appropriate foods, portion control, and foodbudget guidelines. Prescribed energy intakes ranged from5021 to 6276 kJ/d. Energy calculations for weight loss wereobtained by subtracting 2092 kJ from the Harris-Benedictequations using adjusted body weight [31]. Mothers had theoption of lowering their energy intake, increasing theirphysical activity levels, or integrating both strategies. Avariety of resources were offered to participants, includingexchange lists, calorie and fat counters, detailed menus,walking plans, and strength training routines.

Behavioral aspects of weight management presentedincluded social support, stimulus control, contingencymanagement, and stress management [32]. The third part ofthe curriculum, physical activity, consisted of 30-minute in-class exercise sessions and class discussions as described byClarke et al [33]. In addition, these recommendations for diet,physical activity, and behavior modification were tailored toeach participant by providing feedback on diet recalls,pedometer records, and behavior modification worksheets.

The curriculum was based on the social cognitive theoryby Bandura [34]. This theory proposes that dynamic andreciprocal interactions between cognitions, behaviors, andthe environment determine the actions of individuals.Specifically, the intervention promoted a healthier environ-ment (increased availability of healthful foods and supportfor regular physical activity) by affecting mothers' cogni-tions and behaviors. Mothers' cognitions were altered bysetting weight loss goals and increasing knowledge of theenergy content of foods, nutrition label guidelines, meal

planning, and recipe modification. Behavioral changesincluded increased self-efficacy for healthful eating andexercise, enhanced social support by other mothers,increased self-control to manage problem eating and exercisebehaviors, and role modeling for observational learning.Different constructs of the theory were addressed in eachclass. For example, the opening class strived to raiseawareness of the benefits of weight loss and healthful eating(outcome expectations); the cooking demonstration classaimed to enhance the knowledge and skills for mealpreparation (behavioral capability); and the stress manage-ment class focused on the strategies to control binge eating(emotional coping responses). Additional information on theclasses is detailed in Klohe-Lehman et al [35].

2.4. Anthropometric measurements

Height was determined with a stadiometer (PerspectiveEnterprises, Portage, MI), and weight was measured with anelectronic weighing scale (Model HS-100-A, FairbanksScales, St Johnsbury, VT). Body mass index was calculatedas kilogram per square meter. Percentage of body fat wasassessed via bioimpedance with a body compositionanalyzer (Model TBF-300A, Tanita Corporation, ArlingtonHeights, IL). Waist circumference was obtained by position-ing a measuring tape around the abdomen at the highestlateral border of the right iliac crest, as recommended byNHANES III [36].

2.5. Dietary assessment

Intervention mothers reported a 24-hour diet recall and2 days of food records at weeks 0 and 8, and a 24-hour recallweekly during each class. The weekly 24-hour recalls wereconducted for self-monitoring to provide feedback on dietaryintakes. Subjects were provided with oral and writteninstructions by registered dietitians who used measuringspoons and cups for portion size estimation. Nutrient datawere analyzed via Food Processor 7.81 (2001, ESHAResearch, Salem, OR). Nutrient values were modified withthe Software for Intake Distribution Estimation (Side,version 1.0, 2002, Ames, IA). This software corrected forthe variability in daily consumption and produced adjustedestimates of usual intake distributions.

2.6. Survey methods

The Nutrition Attitudes Scale consisted of 21 items with4 subscales—sensory motivators, emotional eating, per-ceived barriers, and healthful eating [27]. The sensorysubscale encompassed taste, hunger, and cravings asmotivators for eating. The emotional eating subscaleassessed eating in response to states of depression, stress/anxiety, and anger. Perceived barriers included dislike oflow-fat foods, less healthful family preferences, confusionregarding nutrition guidelines, too much effort, employ-ment, and lack of interest in changing habits. The healthfuleating subscale incorporated items pertaining to theenjoyment and importance of nutritious foods. Items were

16 K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

evaluated according to a Likert scale ranging from 1 = leastimportant to 7 = very important (sensory motivators,perceived barriers, healthful eating) or 1 = never to 7 =always (emotional eating responses). Each subscale scoreconsisted of the average sum of items in the subscale.Higher scores represented more of the measured trait.Factor analysis was used to demonstrate validity of thesubscales. Reliability of the subscales varied from fair forsensory motivators (α = .65) to very good for promoters ofhealthful eating (α = .86). In addition, demographic datawere elicited with a 40-item questionnaire adapted fromWalker et al [37].

2.7. Statistical methods

The SPSS software (version 11.5, 2003, SPSS Inc,Chicago, IL) was used to analyze data. Entered data werechecked for accuracy, scanned for missing values, evaluatedfor the presence of outliers, and assessed for normality ofdistribution. Analyses were conducted with consideration ofall available data. Statistical significance was shown only ifthe probability (P) values were less than .05.

Table 1Baseline demographic profile of participants enrolled in a nutrition and physical acategory vs nonintervention comparison mothers

Characteristic Intervention grou

Responders (n = 60)

n %

Age (y)18-24 20 33.325-29 20 33.330-39 17 28.4≥40 3 5.0

BMI (kg/m2)25.0-29.9 12 20.030.0-34.9 23 38.335.0-39.9 16 26.7≥40.0 9 15.0

Educational levelHigh school not completed 14 23.3High school graduate 18 30.0Partial college 21 35.0College/graduate degree 7 11.7

EmployedYes 49 81.7No 11 18.3

Children in household1-2 42 70.03-4 15 25.0≥5 3 5.0

Living with spouse/partner*Yes 50 83.3No 10 16.7

Responders significantly different from nonresponders and overall intervention ggroup at *Pb .05 as determined by the χ2 test.

Baseline differences on demographics and anthropo-metrics between groups (intervention and comparison,intervention and program dropouts, intervention respondersand nonresponders) were assessed using the χ2 or Fisherexact test and independent-sample t test [38]. Changes inanthropometrics, attitudes, and nutrients for interventionsubjects were evaluated with the paired t and Wilcoxonsigned rank tests. Differences in anthropometrics and attitudescores between weight loss responders and nonresponders atweek 8 were assessed with analysis of covariance, adjustingfor pretest values. Relationships between weight loss andcontinuous and categorical variables were tested withPearson and Spearman correlation coefficients, respectively.

3. Results

The demographic profile of intervention subjects byweight loss responder category vs nonintervention compar-ison mothers is shown in Table 1. Overall, the mean ages ofthe intervention responders, nonresponders, and comparisonmothers were 27.5, 27.0, and 30.8 years, respectively. The

ctivity intervention for low-income mothers based on weight loss responder

p (n = 114) Comparison group(n = 33)

Nonresponders (n = 54)

n % n %

18 33.3 10 30.323 42.6 3 9.19 16.7 15 45.44 7.4 5 15.2

12 22.2 11 33.319 35.2 7 21.210 18.5 10 30.313 24.1 5 15.2

8 14.8 8 24.216 29.6 11 33.321 38.9 9 27.39 16.7 5 15.2

44 81.5 28 84.810 18.5 5 15.2

37 68.5 17 51.516 29.6 14 42.41 1.9 2 6.1

34 63.0 18 54.520 37.0 15 45.5

roup (responders + nonresponders) significantly different from comparison

Table 2Anthropometric measurements of participants enrolled in a nutrition and physical activity intervention for low-income mothers based on weight loss respondercategory vs nonintervention comparison mothers

Variable Intervention group (n = 114) Comparison group (n = 33)

Responders (n = 60) Nonresponders (n = 54)

Week 0 Week 8 Week 0 Week 8 Week 0 Week 8

Weight (kg) 91.9 ± 18.9 87.2 ± 19.0a 92.1 ± 21.5 91.6 ± 21.4a 90.4 ± 19.4 90.5 ± 19.8a

Body fat (%) 43.1 ± 5.7 41.3 ± 6.3a 43.0 ± 6.0 42.5 ± 6.0a 43.2 ± 5.3 42.9 ± 6.4Waist circumference (cm) 106.1 ± 16.9 101.2 ± 16.6b 108.2 ± 18.5 106.2 ± 18.3b 106.9 ± 14.2 106.9 ± 13.7b

Common superscripted letter (a) within a row indicates significant differences, with responders significantly different than nonresponders and overallintervention group (responders + nonresponders) significantly different than comparison group for week 8 data at P b .001 as determined by analysis ofcovariance, adjusting for week 0 value. Common superscripted letter (b) within a row indicates significant differences, with responders significantly differentthan nonresponders and overall intervention group (responders + nonresponders) significantly different than comparison group at P b .01.

17K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

average BMI differed little for the responder, nonresponder,and overweight comparison groups (34.7, 35.3, and 34.4) anddid not vary by ethnicity. Among responders, 63% wereHispanic, 14% were African American, and 23% were white,as compared to 65%, 26%, and 9% for nonresponders, and49%, 39%, and 12% for comparison mothers, respectively.Themajority of subjects were employed, graduated from highschool, and had 1 to 2 children living at home.More than 80%of responders were living with a partner as opposed to 63% ofthe nonresponders (P b .05). Furthermore, the overallintervention sample of responders and nonresponders wasmore likely to cohabitate with a partner than the comparisongroup (74% vs 54%, P b .05). Most responders (97%),nonresponders (100%), and comparison mothers (85%)qualified for WIC services and reported an annual householdincome less than $29,999 (72%, 79%, and 62%, respectively).

The anthropometric changes for the intervention group byresponder category vs nonintervention comparison mothersare displayed in Table 2. Initially, the mean body weight forresponders, nonresponders, and comparison mothers was91.9, 92.1, and 90.4 kg, with similarities in body fat (43%)and waist circumference measures (range, 106-108 cm). As aresult of the 8-week program, responders significantlydecreased weight (x = −4.7 kg; median, −8.8 kg; P b

Fig. 1. Motivations for eating and nutritional attitudes of intervention mothers by wError bars represent SDs. Responders significantly differ from nonresponders for wby analysis of covariance, adjusting for week 0 score.

.001), body fat (x = −1.8%; P b .001), and waistcircumference (x = −4.9 cm; P b .001) to a greater extentthan nonresponders (weight [x = −0.5 kg; median = −1.4 kg;P b .05], body fat [x = −0.5%], and waist circumference[x = −2.0 cm; P b .05]). For the overall intervention sample,the declines in body weight (x = −2.7 kg; median = −2.4 kg;P b .001), body fat (x = −1.2%; P b .001), and waistcircumference (x = −3.5 cm; P b .001) were highlysignificant. Ninety-eight participants (86%) lost weight;1 person (1%) maintained the same weight; 15 individuals(13%) gained weight. Furthermore, for intervention subjectsavailable at follow-up (week 24), the declines in body weight(x = −2.7 kg; P b .001), percentage of body fat (−0.8%; P b.01), and waist circumference (−12.1 cm; P b .001) remainedsignificantly lower than baseline. In contrast, the overweightcomparison mothers gained a slight, but nonsignificant,amount of weight during the 8-week intervention period(x = 0.1 kg).

Fig. 1 displays the changes in motivators for eating andnutritional attitudes for participants. At baseline, the sensorymotivators were rated as greater influences on eating thanwere attitudes regarding health, emotions, and barriers.However, healthful attitudes ranked higher than sensorymotivators by the end of the program. Subjects reported

eight loss responders (n = 60) and nonresponders (n = 54) at weeks 0 and 8.eek 8 values at P b .01 (double asterisk) and P b .05 (asterisk) as determined

18 K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

more favorable attitudes toward healthful eating, reducedemotional eating, and fewer perceived barriers at postinter-vention than at baseline. In addition, differences in attitudeswere observed by weight loss responder categories. Thosewho lost at least 2.27 kg reported a greater emphasis onhealthful eating attitudes and fewer perceived barriers thannonresponders (b2.27 kg).

The subscale of healthful eating attitudes represented itemsthat were not easily comparable; however, all items improvedas a result of the intervention. Examples of these healthfuleating questions included “I enjoy low-fat foods” and “I amvery concerned with the nutritional content of foods.”With thesubscale of emotional eating, intervention participants asso-ciated depression and stress/anxiety more often with over-eating (78% and 70%, respectively) than anger (59%). Bypostintervention, women were less likely to eat more whendepressed (63%) or angry (56%); however, overeatingassociated with stress exhibited little change (68%). For theperceived barrier subscale, subjects initially ranked the dislikeof low-fat foods (68%), confusion regarding nutrition (68%),and family preferences (58%) highest among barriers. Theseall decreased upon program completion.

Several improvements in dietary intakes were related tochanges in perceived barriers to healthful eating. A greaterconsumption of dairy foods was negatively related toperceived barriers (r = −0.22; P b .05). In addition, motherswho increased their dairy servings reported less confusionregarding nutrition (r = −0.28; P b .01) and fewer complaintsof the effort required to eat healthful foods (r = −0.26; P b.01). Also, subjects who decreased their cholesterol intake bypostintervention reported less difficulty in changing theirdietary habits (r = 0.24; P b .01). Further information onthe dietary results of the intervention was detailed in Clarkeet al [39].

Overall, the weight loss intervention was rated highly.More than 90% of participants reported learning a great dealfrom the program. In particular, women stated that the in-class exercise (88.3%), weekly weigh-ins (85.3%), andwearing a pedometer (84.7%) were very useful components.Suggestions for future classes included establishing main-tenance classes, offering more cooking classes, and provid-ing the program in Spanish.

4. Discussion

Few studies have assessed the impact of a nutrition andphysical activity intervention on weight loss while examin-ing nutrition attitudes in a population of low-incomemothers. Our intervention produced significant reductionsin weight, percentage of body fat, and waist circumference.These reductions are noteworthy because weight loss is verychallenging, especially for low-income populations. Themean weight loss at 8 weeks was comparable to other studiesinvolving minorities such as Kanders et al [7] (2.9 kg) andKumanyika and Charleston [13] (2.7 kg). Weight loss washigher in this study than that for Domel et al [12] (1.4 kg),

Mayer-Davis et al [15] (1.2 kg), and Sullivan and Carter [40](0.2 kg) and lower than that for Ard et al [11] (6.7 kg), Kauland Nidiry [9] (6.4 kg), and McNabb et al [10] (4.4 kg). Thatmothers were able to maintain significant decreases inweight, percentage of body fat, and waist circumference for6 months shows that this program has potential forpromoting lasting health benefits in this population.

This study is unique for its comparison of sensory andbehavioral attitudes of nutrient intake by low-incomewomen. At baseline, mothers reported a greater influenceof sensory motivators, such as taste and hunger, on eatingchoices rather than specific attitudes. It is well establishedthat taste is a major reason for food selection [41]. However,our sensory subscale did not change as a result of theintervention. This lack of change was not unexpectedbecause sensory factors are largely influenced by biologicalmechanisms [42]. Upon program completion, healthfuleating concerns superseded sensory attributes as the mostimportant motivator for eating.

The reduction in emotional eating and perceived barriersat the end of the intervention corroborate other enhance-ments in attitudes by mothers enrolled in studies targetingcancer prevention [23,24], dietary fat reduction [21,22], andfruit and vegetable consumption [25,26]. In contrast,Fitzgibbon et al [19] did not find significant changes innutrition attitudes in an obesity prevention study for AfricanAmericans. The authors attributed this result to the smallsample size of 24 participants and short treatment durationof 6 weeks.

The improvements in perceived barriers toward healthfuleating are believed to have contributed to the positive dietarychanges observed in this study. The associations betweenattitudes and dietary behavior do not prove a cause and effectrelationship but support the proposition by Bandura [34] thatcognitive changes interact with behavior.

Limitations of this study include its high attrition andshort treatment period. Yet, the attrition rate for this program(56%) is within the range of 23% to 80% experienced inother studies recruiting minorities [9,10,12,13,15,43]. Fac-tors influencing attrition in this program included illness of achild, lack of childcare, transportation difficulties, jobconflicts, financial constraints, family responsibilities, insuf-ficient time, lack of family support, personal stress, andrespondent burden of the questionnaires. These barriers toparticipation precluded a longer intervention and have beendocumented in previous research with low-income women[8]. Although we attempted to minimize these issues byproviding a free program, offering makeup classes, andvarying class times and locations, obstacles remained.

In summary, a curriculum was developed to promoteweight loss in a population of low-income mothers.Significant declines in weight, percentage of body fat, andwaist circumference were observed. Furthermore, we foundthat weight loss responders improved their attitudes towardhealthful eating and reduced their perceived barriers tohealthful eating to a greater extent than nonresponders by

19K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

postintervention. These results demonstrate that a lifestyleprogram can promote weight loss in low-income mothers.Therefore, public health clinics should consider adoptingweight management programs for their clients.

Acknowledgment

This research was supported by a grant from theTexas Higher Education Coordinating Board, Austin, TX(UTA#00-377).

References

[1] Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, FlegalKM. Prevalence of overweight and obesity among US children,adolescents, and adults, 1999-2002. JAMA 2004;291:2847-50.

[2] Cossrow N, Falkner B. Race/ethnic issues in obesity and obesity-related comorbidities. J Clin Endocrinol Metab 2004;89:2590-4.

[3] Zang Q, Wang Y. Socioeconomic inequality of obesity in the UnitedStates: Do gender, age, and ethnicity matter? Soc Sci Med 2004;58:1171-80.

[4] Visscher TL, Seidell JC. The public health impact of obesity. Annu RevPublic Health 2001;22:355-75.

[5] Cogan JC, Rothblum ED. Outcomes of weight-loss programs. GenetSoc Gen Psychol Monogr 1992;118:385-415.

[6] Braveman P, Egerter S, Marchi K. The prevalence of low incomeamong childbearing women in California: implications for the privateand public sectors. Am J Public Health 1999;89:868-74.

[7] Kanders BS, Ullmann-Joy P, Foreyt JP, Heymsfield SB, Heber D,Elashoff RM, et al. The black American lifestyle intervention (BALI):the design of a weight loss program for working-class African-American women. J Am Diet Assoc 1994;94:310-2.

[8] French SA, Jeffery RW, Story M, Neumark-Sztainer D. Perceivedbarriers to and incentives for participation in a weight-loss programamong low-income women in WIC. J Am Diet Assoc 1998;98:79-81.

[9] Kaul L, Nidiry JJ. Management of obesity in low-income AfricanAmericans. J Natl Med Assoc 1999;91:139-43.

[10] McNabbWL, QuinnMT, Rosing L. Weight loss program for inner-cityblack women with non–insulin-dependent diabetes mellitus: PATH-WAYS. J Am Diet Assoc 1993;93:75-7.

[11] Ard JD, Rosati R, Oddone EZ. Culturally-sensitive weight lossprogram produces significant reduction in weight, blood pressure, andcholesterol in eight weeks. J Natl Med Assoc 2000;92:515-23.

[12] Domel SB, Alford BB, Cattlett HN, Gench BE. Weight control forblack women. J Am Diet Assoc 1992;92:346-448.

[13] Kumanyika SK, Charleston JB. Lose weight and win: a church-basedweight loss program for blood pressure control among black women.Patient Educ Couns 1992;19:19-32.

[14] Agurs-Collins TD, Kumanyika SK, Ten Have TR, Adams-CampbellLL. A randomized controlled trial of weight reduction and exercise fordiabetes management in older African-American subjects. DiabetesCare 1997;20:1503-11.

[15] Mayer-Davis EJ, D'Antonio A, Martin M, Wandersman A, Parra-Medina D, Schultz R. Pilot study of strategies for effective weightmanagement in type 2 diabetes: Pounds Off with Empowerment(POWER). Fam Community Health 2001;24:27-35.

[16] Pinto BM, Clark MM, Cruess DG, Szymanski L, Pera V. Changes inself-efficacy and decisional balance for exercise among obese womenin a weight management program. Obes Res 1999;7:288-92.

[17] Axelson M, Federline T, Brinberg DA. A meta-analysis of food andnutrition-related research. J Nutr Educ 1985;17:51-4.

[18] Story M, Neumark-Sztainer D, French S. Individual and environmentalinfluences on adolescent eating behaviors. J Am Diet Assoc 2002;102(Suppl 3):S40-S51.

[19] FitzgibbonML, Stolley MR, Kirschenbaum DS. An obesity preventionpilot program for African-American mothers and daughters. J NutrEduc 1995;27:93-9.

[20] Clarke KK, Freeland-Graves J, Klohe-Lehman DM, Bohman TM.Predictors of weight loss in low-income mothers of young children.J Am Diet Assoc 2007;107:1146-54.

[21] Albright CL, Bruce B, Howard-Pitney B, Winkleby MA, Fortmann SP.Development of a curriculum to lower dietary fat intake in a multiethnicpopulation with low literacy skills. J Nutr Educ 1997;29:215-23.

[22] Howard-Pitney B, Winkleby MA, Albright CL, Bruce B, Fortmann SP.The Stanford Nutrition Action Program: a dietary fat intervention forlow-literacy adults. Am J Public Health 1997;87:1971-6.

[23] Cox RH, Carmen M, Gonzales-Vigilar RV, Novascone MA, Silva-Barbeau I. Impact of a cancer intervention on diet-related cardiovas-cular disease risks of white and African-American EFNEP clients.J Nutr Educ 1996;28:209-18.

[24] Harnack L, Block G, Subar A, Lane S, Brand R. Association of cancerprevention-related nutrition knowledge, beliefs, and attitudes to cancerprevention dietary behavior. J Am Diet Assoc 1997;97:957-65.

[25] Havas S, Damron D, Treiman K, Anliker J, Langenberg P, HammadTA, et al. The Maryland WIC 5 a day program pilot study: rationale,results, and lesson learned. J Nutr Educ 1997;29:343-50.

[26] Treiman K, Freimuth V, Damron D, Lasswell A, Anliker J, Havas S,et al. Attitudes and behaviors related to fruits and vegetablesamong low-income women in the WIC program. J Nutr Educ 1996;28:149-56.

[27] Nuss H, Clarke K, Klohe-Lehman D, Freeland-Graves J. Influence ofnutrition attitudes andmotivators for eating on postpartumweight statusin low-income new mothers. J Am Diet Assoc 2006;106:1774-82.

[28] U.S. Department of Health and Human Services. The SurgeonGeneral's call to action to prevent and decrease overweight andobesity, 2001. Available at: http://www.surgeongeneral.gov/topics/obesity/calltoaction/cover.htm Accessed, September 22, 2007.

[29] U.S. Departments of Agriculture and Health and Human Services.Nutrition and your health: Dietary Guidelines for Americans. 5th ed.Washington, DC: Home and Garden Bulletin No. 232; 2000.

[30] Johnson RK, Kennedy E. The 2000 Dietary Guidelines for Americans:what are the changes and why were they made? J Am Diet Assoc 2000;100:769-74.

[31] Frankenfield DC, Rowe WA, Smith JS, Cooney RN. Validation ofseveral established equations for resting metabolic rate in obese andnonobese people. J Am Diet Assoc 2003;103:1152-9.

[32] Thomas PR. Weighing the options: criteria for evaluating weightmanagement programs. Washington, DC: National Academy Press;1995.

[33] Clarke KK, Freeland-Graves J, Klohe-Lehman DM, Milani TJ, NussHJ, Laffrey S. Promotion of physical activity in low-income mothersusing pedometers. J Am Diet Assoc 2007;107:962-7.

[34] Bandura A. Social foundations of thought and action: a social cognitivetheory. Englewood Cliffs, NJ: Prentice Hall PTR; 1986.

[35] Klohe-Lehman DM, Freeland-Graves J, Anderson ER, McDowell T,Clarke KK, Hanss-Nuss H, et al. Nutrition knowledge is associatedwith greater weight loss in obese and overweight low-income mothers.J Am Diet Assoc 2006;106:65-75.

[36] National Heart, Lung, and Blood Institute. Clinical guidelines on theidentification, evaluation, and treatment of overweight and obesity inadults: the evidence report, NIH Publication No. 98-4083, September1998.

[37] Walker LO, Freeland-Graves JH, Milani T, Hanss-Nuss H, George G,Sterling BS, et al. Weight and behavioral and psychosocial factorsamong ethnically diverse, low-income women after childbirth:I. methods and context. Women Health 2004;40:1-17.

[38] Munro BH. Statistical Methods for Health Care Research. New York,NY: Lippincott; 1997.

[39] Clarke KK, Freeland-Graves J, Klohe-Lehman DM, Bohman TM.Predictors of weight loss in low-income mothers of young children.J Am Diet Assoc 2007;107:1146-54.

20 K.C. Jordan et al. / Nutrition Research 28 (2008) 13–20

[40] Sullivan J, Carter JP. A nutrition-physical fitness intervention programfor low-income black parents. J Natl Med Assoc 1985;77:39-43.

[41] Freeland-Graves J, Nitzke S. Position of the American DieteticAssociation. Total diet approach to communicating food and nutritioninformation. J Am Diet Assoc 2002;102:100-8.

[42] Yamamoto T, Nagai T, Shimura T, Yasoshima Y. Roles of chemicalmediators in the taste system. Jpn J Pharmacol 1998;76:325-48.

[43] Stevens VJ, Corrigan SA, Obarzanek E, Bernauer E, Cook NR, HebertP, et al. Weight loss intervention in phase I of the Trials ofHypertension Prevention. Arch Intern Med 1993;153:849-58.