Stress Management–Augmented Behavioral Weight Loss Intervention for African American Women

Abstract

The relationship between chronic stress and weight management efforts may be a concern for African American (AA) women, who have a high prevalence of obesity, high stress levels, and modest response to obesity treatment. This pilot study randomly assigned 44 overweight/obese AA women with moderate to high stress levels to either a 12-week adaptation of the Diabetes Prevention Program Lifestyle Balance intervention augmented with stress management strategies (Lifestyle + Stress) or Lifestyle Alone. A trend toward greater percentage of baseline weight loss at 3-month data collection was observed in Lifestyle + Stress (−2.7 ± 3.6%) compared with Lifestyle Alone (−1.4 ± 2.3%; p = .17) and a greater reduction in salivary cortisol (Lifestyle + Stress: −0.2461 ± 0.3985 ng/mL; Lifestyle Alone: −0.0002 ± 0.6275 ng/mL; p = .20). These promising results suggest that augmenting a behavioral weight control intervention with stress management components may be beneficial for overweight/obese AA women with moderate to high stress levels and merit further investigation with an adequately powered trial.

Keywords

African American obesity stress weight women

The obesity epidemic is disproportionately represented among minority populations, with highest rates among African American women (Flegal, Carroll, Ogden, & Curtin, 2010; Ogden et al., 2006). Among adult women assessed in the 2007-2008 National Health and Nutrition Examination Survey, ~50% of African Americans were obese compared with 43% for Hispanics and 33% for Whites (Flegal et al., 2010). Furthermore, African Americans have a high likelihood of developing obesity-related comorbidities such as diabetes and cardiovascular disease (Kumanyika, 1994; Stevens, 2000). The combination of higher obesity prevalence and highly prevalent obesity-related conditions that contribute to elevated morbidity and mortality has been suggested as one of the pathways that contribute to known health disparities between African Americans and Caucasians (McTigue et al., 2006; Must et al., 1999; Wang & Beydoun, 2007). To further exacerbate these health disparities, it appears that current behavioral obesity treatment approaches produce significantly lower weight losses among African Americans than observed for Caucasians, particularly among African American women (Kumanyika et al., 2002; West, Prewitt, Bursac, & Felix, 2008). Methods to improve the weight losses of overweight African American women are a public health imperative and calls to action have been issued (Ard et al., 2008; Kim, Bursac, DiLillo, White, & West, 2009; Kumanyika et al., 2005).

African American women report higher rates of chronic stress (Turner & Avison, 2003) and stressful life events (Schulz et al., 2000; Turner & Avison, 2003) compared with non-Hispanic Caucasians. Mechanisms by which elevated levels of chronic stress might be positively associated with obesity include both physiologic and behavioral avenues. African Americans display higher physiological measures of stress, such as cortisol, than Caucasians, even after controlling for income (Geronimus, Hicken, Keene, & Bound, 2006). Elevated cortisol levels are characteristic both of higher stress levels and of increased eating behavior and have been associated with higher body weight (Farag et al., 2008; Vicennati, Pasqui, Cavazza, Pagotto, & Pasquali, 2009). Stress-induced elevations in cortisol can directly favor fat accumulation, thereby indirectly promoting weight gain, by inhibiting lipolysis or by inhibiting growth hormone (Bjorntorp, 2001; Bjorntorp & Rosmond, 2000). Stress may also affect effective weight control through behavioral avenues, such as eating behavior related to stress (Economos, Hildebrandt, & Hyatt, 2008; Manzoni et al., 2009). Indeed, chronic stress has been shown to be associated with unsuccessful weight loss efforts (Delahanty, Meigs, Hayden, Williamson, & Nathan, 2002; Smith & McFall, 2005; Wadden et al., 2006; Walcott-McQuigg, 1995) and vulnerability to weight gain (Economos et al., 2008). Researchers examining the relationships between stress, eating behavior, and obesity have called for the inclusion of physiological measures of stress in addition to self-report, with a specific recommendation that cortisol be collected (Torres & Nowson, 2007).

Taken together, there is a growing body of research demonstrating that stress could undermine effective weight management efforts and suggesting that the introduction of stress management strategies may confer particular advantages for overweight African American women engaged in behavioral obesity interventions since they are more likely to experience chronic stress (Turner & Avison, 2003) and are at risk of attenuated success in behavioral weight loss programs (Kumanyika et al., 2002; West et al., 2008). Although comprehensive behavioral weight management programs often touch on emotional coping and other stress management strategies, these are not a primary focus of most empirically validated behavioral lifestyle interventions (Diabetes Prevention Program Research Group, 2002; Wadden et al., 2006). Nonetheless, there is some indication that relaxation training can be effective in reducing emotional eating episodes in obese women (Manzoni et al., 2009), although the impact on weight loss was not examined and the sample did not include African American women. In addition, recent qualitative work indicates that a stronger emphasis on stress management within the context of weight management has appeal for African American women (Cox, Zunker, Wingo, Jefferson, & Ard, 2011).

Although these studies support a benefit of stress management and coping skills as part of a comprehensive weight loss and/or maintenance program, they do not specifically address whether a strong emphasis on stress management techniques among individuals with chronic stress improves weight loss achieved in a behavioral weight control program or enhances treatment engagement. Furthermore, none of the intervention studies have specifically focused on overweight African American women, a group that appears to be particularly vulnerable to chronic stress. Therefore, this randomized, controlled pilot study evaluated a novel stress management–augmented 12-session group-based behavioral weight control program in comparison with a traditional behavioral lifestyle program in this high-risk population. The specific hypotheses examined were that the stress management–augmented program would produce greater weight loss and reductions in stress level, as well as greater program adherence, than the standard behavioral approach.

Materials and Method

Participants

Overweight and obese African American women aged 21 years and older with a body mass index (BMI; calculated as weight [kg]/height [m²]) between 25 and 45 were recruited over a 4-week period in January 2010 by word of mouth and emails distributed to local organizations (i.e., workplaces, churches, community groups) and previous research participants. To be eligible, participants were required to be female, self-identify as African American, and report an elevated stress level on Cohen’s Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983). A score of 17 or higher was considered eligible for participation as this score corresponds to moderate and high levels of perceived stress in women presenting for obesity treatment in previous research (Kim et al., 2009). Women were also required to be free of medical problems that would contraindicate participation in a behavioral weight loss program containing an exercise component (e.g., heart attack or stroke in the previous 6 months, insulin-dependent diabetes, physical inability to be active), be willing to be randomized to either of the treatment conditions, and have no plans to move from the area for the study period. Exclusion criteria included current pregnancy or lactation, weight loss of 10% or more in the previous 6 months, previous bariatric surgery, or current participation in another formal weight reduction program. The study was approved by the institutional review board at the University of Arkansas for Medical Sciences.

Dependent Measures

All measures were obtained at baseline screening (January 2010) and at postintervention data collection (May 2010) unless otherwise indicated. Sociodemographic characteristics such as age, education, self-reported race and ethnicity, and marital status, as well as health history, were collected by self-report at baseline.

Body weight

Weight was measured in indoor clothing, without shoes, on a calibrated digital scale (Tanita, BWB 800, Tokyo, Japan). Height was measured using a wall-mounted stadiometer (Seca Corporation, Hanover, MD). BMI was calculated as weight (kg)/height (m²).

Stress levels

Participant stress levels were assessed by self-report using Cohen’s PSS (Cohen et al., 1983), a 14-item inventory that has been widely used in a variety of populations and for numerous health conditions. The measure assesses the degree to which situations in one’s life in the previous month are appraised as stressful; scores range from 0 to 56, with higher scores on the PSS indicating greater perceived stress. An example of one question on the survey is, “In the last month, how often have you found that you could not cope with all the things that you had to do?” Each question has five response options, which are never, almost never, sometimes, fairly often, and very often. These response options are coded on a 0 to 4 scale, which is then used to calculate the PSS score. The PSS has been validated, found to have adequate internal test–retest reliability (Cronbach’s α = .88), and is positively correlated with a variety of self-report and behavioral indices of stress in adult populations (Cohen et al., 1983). Additionally, the PSS has been demonstrated to be appropriate for use in diverse populations including African American women and has been used in previous research studies examining stress in African Americans (Cole, 1999; Fowler-Brown et al., 2009; Sharp, Kimmel, Kee, Saltoun, & Chih-Hung, 2007).

Salivary cortisol was also collected to provide a physiological measure of stress. Participants were instructed to collect their saliva after they woke up in the morning, but before they had breakfast or consumed any liquids, by emitting saliva (i.e., spitting) by mouth into a saliva collection tube (ALPCO #B-SCD) with a cover. If the participant was unable to emit saliva into the tube, they were instructed to insert a cotton roll between the cheek and gum and place the cotton roll in the tube. The saliva tube was then stored in their home freezer. The tubes were prelabeled with the participant’s study ID number and a space for the participant to record the collection date. Participants collected and stored saliva in their home freezer for 3 consecutive mornings prior to their data collection visits (baseline and follow-up) at which time they brought the tubes in to the research lab. All salivary cortisol specimens were analyzed at once using an ALPCO Diagnostic salivary cortisol EIA kit (Salem, NH) and following the provided protocol. The cortisol values from each of the 3 days of collection within a single assessment displayed skewed distributions, which is common especially when saliva is collected in the early morning hours (Luecken & Gallo, 2007). Therefore, mean cortisol values (ng/mL; mean of the 3 sequential days of data collection) were logarithmically transformed to improve the likelihood of approximating a normal distribution to allow analyses for which normality is an underlying assumption. Both baseline and follow-up salivary cortisol values were log-transformed for analyses.

Treatment adherence and satisfaction

Adherence data collected included attendance at group sessions and submission of weekly self-monitoring diaries. Program satisfaction was obtained via questionnaire at posttreatment only.

Behavioral Lifestyle Intervention Conditions

Both treatment conditions targeted dietary and physical activity habit change and provided instruction in behavioral strategies to assist in effectively modifying eating and exercise behaviors. Both conditions were offered 12 weekly group sessions that lasted approximately 60 minutes and were delivered by one of two African American masters-level counselors, that is, one counselor delivered the Lifestyle Alone sessions and a different counselor delivered the Lifestyle + Stress sessions. Both counselors were trained in health behavior/health promotion and have expertise in the delivery of behavioral weight loss interventions.

Program content was based on the Diabetes Prevention Program (Diabetes Prevention Program Research Group, 2002) and the Look AHEAD behavioral lifestyle program (Look AHEAD Research Group, 2006). Participants received calorie and fat intake goals for inducing weight loss tailored to their starting weight, graded physical activity goals with an eventual goal of 200 minutes/week of physical activity, and a personal weight loss goal of 10% or more of initial body weight. Graded physical activity goals encouraged participants to work up to 60 minutes/week of moderate-to-vigorous physical activity by Week 4 and to increase by 30 minutes each week thereafter, to reach the treatment goal of 200 minutes/week by Week 9. Self-monitoring diaries were provided weekly to record dietary intake and physical activity; counselors reviewed diaries weekly and provided feedback on progress toward goals and behavior changes made. Participants assigned to the standard lifestyle (Lifestyle Alone) program were introduced to some stress management concepts and strategies commonly included in behavioral weight management programs (e.g., problem-solving skills). Both interventions were protocol-driven, with session content outlined for each session of each study arm. Treatment fidelity was monitored in weekly supervision sessions with senior study investigators and was based on in vivo observation of components delivered in each session within each arm by research assistants.

Stress Management–Augmented Lifestyle Intervention

The novel stress management–augmented lifestyle intervention incorporated all the components of the standard lifestyle program as well as stress management strategies (Lifestyle + Stress). Each week a different stress management technique was introduced to participants, with in-class practice and instructions to self-monitor stress management techniques used during the upcoming week. Participants were also asked to provide periodic ratings of their stress level during the day as well as before and after stress management technique rehearsal. The stress management approaches included relaxation (e.g., guided relaxation, diaphragmatic breathing, mindfulness), behavioral approaches (e.g., distraction, time management, gum chewing), and cognitive strategies (e.g., challenging dichotomous thinking, positive self-statements, humor), which are described in more detail below. The goal of this “sampler” of stress management techniques was to introduce participants to a range of evidence-based stress reduction methods that spanned different domains and ask them to practice the different techniques to determine which were most useful for them in various situations, allowing personalization of the stress management strategies.

Relaxation methods

Participants were given instruction in diaphragmatic breathing, progressive muscle relaxation, and various types of guided imagery methods. In addition to in-session practice, all participants in this condition were provided a CD with recordings of 13 different relaxation scripts. Participants were instructed to use these relaxation strategies and to incorporate diaphragmatic breathing with mindful eating.

Behavioral methods

Distraction techniques such as listening to music or talking with friends were offered as a strategy to help take one’s mind off of the stressor. Gum chewing was also suggested as a distraction technique. Additionally, time management was a behavioral strategy introduced as a stress reduction technique.

Cognitive strategies

Methods to identify and challenge dichotomous thinking were introduced. Positive self-statements to affirm achievements, strengths, and assets were also introduced to facilitate maintenance of a balanced perspective.

Randomization

Following baseline data collection, participants were randomized to the Lifestyle Alone or Lifestyle + Stress arm by a study investigator using a random numbers table. Individuals were informed of their randomization status at their first treatment group session.

Statistical Analyses

Between-group differences on baseline characteristics were analyzed using chi-square tests for categorical variables and t tests for continuous variables. Intention-to-treat analyses were used to analyze weight loss, the primary endpoint, with baseline carried forward for individuals missing 3-month posttreatment data (Kaiser, Affuso, Beasley, & Allison, 2011; Shao, Jordan, & Pritchett, 2009; Ware, 2003). Weight change was computed by calculating the difference between baseline and posttreatment weight values. A general linear model, with weight change as the dependent variable and treatment group as the independent variable, was used to determine whether treatment group was a significant predictor of weight loss. Additionally, we entered attendance and a treatment-by-attendance interaction term to test whether a treatment-by-attendance interaction was a significant predictor of weight loss. Secondary outcomes were analyzed using data for all completers of the measure being analyzed (cortisol data analysis: n = 30; PSS data analysis: n = 36). One-way ANOVA was used to test the main effect of treatment group on changes in stress levels (PSS and salivary cortisol). Correlational analyses were used to determine associations among stress, weight change, and adherence parameters. Additional exploratory analyses were conducted in which PSS scores were dichotomized using a median split to examine outcomes among our moderately to highly stressed sample. All analyses were conducted using PASW Statistics 18, Release Version 18.0.0 (SPSS, Inc., Chicago, IL) and a p value of <.05 was considered statistically significant.

Results

Of the 97 women screened, 44 women were enrolled and randomized (Figure 1). Participants were well-educated (70% college graduate or graduate school), obese (93%) African American women who reported moderate to high stress levels (Table 1). Ninety-six percent of participants were employed full-time. There were no differences in initial body weight or sociodemographic characteristics between study conditions, nor did the groups differ in perceived stress levels or salivary cortisol levels at baseline (Table 1).

Figure 1.

CONSORT diagram

Table 1.

Sample Characteristics

	Lifestyle Alone (n = 22)	Lifestyle + Stress (n = 22)	Total (n = 44)
Baseline sociodemographic factors
Age (years)	43.4 ± 10.2	45.7 ± 11.4	44.5 ± 10.8
BMI (kg/m²)	35.8 ± 4.4	37.5 ± 5.2	36.6 ± 4.8
Weight (kg)	92.1 ± 13.8	100.6 ± 14.9	96.2 ± 32.7
Education
High school graduate	1 (4.5)	0 (0)	1 (2.3)
Some college	6 (27.3)	6 (27.3)	12 (27.3)
College graduate	6 (27.3)	7 (31.8)	13 (29.5)
Graduate School	9 (40.9)	9 (40.9)	18 (40.9)
Marital status
Married	10 (45.5)	12 (54.5)	22 (50.0)
Divorced	6 (27.3)	6 (27.3)	12 (27.3)
Separated	2 (9.1)	1 (4.5)	3 (6.8)
Never married/single	4 (18.2)	1 (4.5)	5 (11.4)
Unmarried couple	0 (0)	2 (9.1)	2 (4.5)
Stress measures
Perceived Stress Scale (PSS)	31.2 ± 7.0	30.2 ± 8.3	30.7 ± 7.6
Log of salivary cortisol (ng/mL)	0.7 ± 0.3	0.6 ± 0.3	0.6 ± 0.3
Behavioral adherence measures
Group sessions attended (out of 12)	7.9 ± 3.9	6.9 ± 2.9	7.4 ± 3.4
Diaries submitted (out of 11)	5.9 ± 3.7	6.0 ± 3.1	6.0 ± 3.4

Note. BMI = body mass index. Data are means ± standard deviations or number (%).

Retention and Adherence

Retention rates, defined as participants for whom we obtained at least a measured weight at 3-month data collection, were high (86%), with no difference between conditions. There were no significant differences in the attendance or the self-monitoring diary adherence patterns between the two conditions. The combined sample attended 62% of the group sessions (Lifestyle + Stress, 57.2%; Lifestyle Alone, 65.9%) and submitted 50% of the self-monitoring diaries on average (Lifestyle + Stress, 50.3%; Lifestyle Alone, 49.2%; Table 1).

Sensitivity Analysis

Fourteen participants had some incomplete data at follow-up data collection, particularly the postintervention saliva collection. Table 2 compares sample characteristics of those who completed all study measures (n = 30) with those who did not complete all measures (n = 14). There was no significant difference in age or baseline weight between those who completed all study measures compared with those who did not. Although there was no difference in baseline self-reported stress for completers compared with noncompleters, those who completed all follow-up measures had lower baseline salivary cortisol levels than did noncompleters (p = .02). Also, those who completed all follow-up measures attended more group sessions (p < .001) and submitted more diaries (p < .001) than noncompleters, suggesting that those who completed all study measures were more engaged than those who did not provide saliva for cortisol analysis at 3-month data collection.

Table 2.

Characteristics of Data Collection Completers and Noncompleters

	Noncompleters (n = 14)	Completers (n = 30)	p Value
Baseline sociodemographic factors
Age (years)	41.1 ± 9.3	46.1 ± 11.2	ns
BMI (kg/m²)	37.7 ± 4.6	36.1 ± 5.0	ns
Weight (kg)	92.4 ± 29.9	89.1 ± 28.6	ns
Stress measures
Perceived Stress Scale (PSS)	30.4 ± 7.1	30.8 ± 7.8	ns
Log of salivary cortisol (ng/mL)	0.8 ± 0.2	0.6 ± 0.4	.02
Behavioral adherence measures
Group sessions attended	3.9 ± 2.2	9.0 ± 2.5	<.001
Diaries submitted	2.5 ± 1.7	7.6 ± 2.7	<.001

Note. BMI = body mass index. Data are means ± standard deviations.

Primary Outcome

Participants in both groups weighed significantly less (p < .001) at 3-month data collection compared with baseline (Table 3), indicating that the group-based behavioral obesity program was effective in promoting weight loss. Weight losses at 3 months averaged 2.0 ± 3.1%, with a trend toward greater weight losses by Lifestyle + Stress than by the Lifestyle Alone (p = .17). A general linear model with weight change as the dependent variable and treatment group, total attendance, and group × attendance interaction term as independent predictors revealed group × attendance interaction as a significant predictor of weight loss (p = .01). This result suggests a multiplicative effect of treatment group and session attendance on achieving greater weight losses (Figure 2). Specifically, these data indicate that there was a stronger positive association between group attendance and weight loss in the Lifestyle + Stress group compared with the Lifestyle Alone group.

Table 3.

Change in Weight and Stress

	Lifestyle Alone (n = 22)	Lifestyle + Stress (n = 22)	Total (n = 44)	p Value^a
Weight loss (kg)	1.3 ± 2.1	2.7 ± 3.9	2.0 ± 3.2	.13
Percentage of baseline weight lost	1.4 ± 2.3	2.7 ± 3.6	2.0 ± 3.1	.17
Change in PSS score^b	−6.5 ± 9.8	−5.3 ± 13.3	−5.8 ± 11.7	.77
Change in log cortisol (ng/mL)^c	−0.0002 ± 0.6275	−0.2461 ± 0.3985	−0.1478 ± 0.5074	.20

Note. PSS = Perceived Stress Scale. Data are means ± standard deviations.

Between group test comparing Lifestyle Alone and Lifestyle + Stress.

Missing for 8.

Missing for 14.

Figure 2.

The relationship between weight change and total sessions attended by treatment group

Secondary Outcomes

Overall, women reported lower perceived stress postintervention (p = .01); however, this reduction in perceived stress did not differ between groups (Table 3). Women in the Lifestyle + Stress condition showed a trend toward greater reductions in salivary cortisol than in the Lifestyle Alone condition (p = .20; Table 3).

Correlations Between Weight Change, Adherence, and Stress

Weight loss was significantly associated with total sessions attended (r = .31, p = .04) and total self-monitoring diaries submitted (r = .39, p = .01). Weight change was not linearly associated with change in either of the stress measures, though weight loss was inversely associated with follow-up PSS (r = −.35, p = .03), such that those who lost more weight reported lower follow-up PSS. There was no evidence of an association between any of the adherence measures and either of the stress measures.

Exploratory Analysis

To further explore the role of baseline perceived stress, we dichotomized baseline PSS scores into “higher stress” or “lower stress” categories. These distinctions are relative since our eligibility criteria stipulated that participants have at least moderate perceived stress at baseline. Participants with baseline PSS > 29.5 (sample median value) were classified as “higher stress,” whereas those with baseline PSS ≤ 29.5 were classified as “lower stress.” Overall, those in the higher stress group reported greater reduction in PSS than those in the lower stress group (−11.9 vs. 1.0, p < .001). However, there were no significant differences between stress groups in weight loss or cortisol reduction (data not shown). To explore whether perceived stress level may have interacted with the intervention efficacy, we tested for a treatment-group (Lifestyle Alone vs. Lifestyle + Stress) by baseline PSS-group (high stress vs. low stress) interaction. The treatment-group × baseline PSS-group interaction term emerged as a significant predictor of reduction in PSS (p = .05), such that those in the Lifestyle + Stress group with the highest baseline PSS showed the greatest reduction in perceived stress compared with the other treatment × baseline PSS combination groups; however, this interaction was not a significant predictor of weight loss or reduction in cortisol level.

Discussion

This study is one of the first to evaluate the feasibility and efficacy of delivering a stress management–augmented behavioral weight control program to a group of moderate to highly stressed overweight and obese African American women. Though not adequately powered to detect a statistically significant difference, the findings of this small pilot study suggest that enhancing a standard lifestyle intervention with stress management techniques may lead to improved outcomes in African American women. Although the change in perceived stress did not statistically differ between the Lifestyle + Stress group compared with the Lifestyle Alone group, the Lifestyle + Stress group trended toward greater weight loss and reduction in salivary cortisol level. The limited 12-week treatment duration may not have provided adequate time for full realization of intervention benefits.

The weight losses achieved by both treatment conditions in this study are comparable with or greater than weight losses typically seen in other trials targeting African American women with interventions of similar duration (Ard et al., 2010; Fitzgibbon et al., 2005; Hermann-Engels, Gretebeck, Gretebeck, & Jimenez, 2005; Kanders et al., 1994) and even comparable with those reported in programs with longer duration (Kennedy et al., 2009; Samuel-Hodge et al., 2009). Our findings are consistent with previous studies that have demonstrated, on average, modest weight losses in African American women who complete a standard behavioral weight control intervention (Ard et al., 2010; Banks-Wallace, 2007; Boltri et al., 2008; Fitzgibbon et al., 2005; Hermann-Engels et al., 2005; Hollis et al., 2008; Karanja, Stevens, Hollis, & Kumanyika, 2002; Kennedy et al., 2009; Kumanyika et al., 2009; Samuel-Hodge et al., 2009; West, DiLillo, Bursac, Gore, & Greene, 2007; West et al., 2007). Thus, we feel confident that the Lifestyle Alone program was a strong comparison group against which to measure the additive benefits of the stress management components. That the Lifestyle + Stress group achieved weight losses double those of the Lifestyle Alone group suggests that the stress management component may well have enhanced weight loss in the long run if the treatment period were extended.

Our study did not demonstrate a significant effect of treatment condition on change in PSS, although there was an overall reduction in PSS. Additionally, though the small number of cases per group in our subgroup analyses may limit confidence in the conclusiveness of findings, there was preliminary evidence to suggest that participants with the highest stress levels who were randomized to the Lifestyle + Stress condition achieved greater reduction in perceived stress than did other study participants. This may indicate that high stress (rather than moderate stress) overweight and obese individuals may be the optimal population to target for a stress management–augmented weight control intervention. Furthermore, based on the means and standard deviations of PSS scores found in previous studies of African American women (Schiffman, Graham, Sattely-Miller, & Peterson-Dancy, 2000; Strickland, Giger, Nelson, & Davis, 2007), we can assume that under a normal distribution, approximately 88% of African American women would have a PSS score of at least 17, also suggesting that a higher cutoff for PSS score may be warranted in future interventions. There is evidence to support that behavioral interventions can be successful at improving stress management behaviors (Katzer et al., 2008) and reducing depression and emotional eating in obese women (Manzoni et al., 2009). However, this literature is generally focused on stress-related behaviors rather than weight loss as the primary outcome. Additionally, these interventions are not typically offered in the context of an obesity treatment intervention. Thus, additional research is needed to understand how the addition of a strong stress management component to a behavioral weight loss intervention can be used to improve participants’ perceived stress and cortisol levels concurrent with achieving the primary outcome, weight loss, as well as to understand the paths by which changes in physiological and self-reported stress may affect the weight loss outcomes.

This study has several strengths, including a novel stress management–augmented intervention and the utilization of an objective measure of stress level. Although it has been suggested that stress management may confer particular advantages for overweight African American women in weight management efforts, to our knowledge, this study was the first to test the effect of an obesity intervention specifically enhanced with stress management techniques on weight loss in this population. The multidimensional assessment of stress that included both a biological marker and a subjective measure of perceived stress confers advantages over other studies of how stress interacts with weight loss in obesity treatment that use only a unitary stress assessment approach. This pilot study is limited by a small sample size, which reduced our power to detect statistically significant differences. Follow-up analyses indicated that the effect size detected in the current study would have reached statistical significance in a sample of size of approximately 180, which would be more typical in a full-scale trial. Furthermore, there was a reasonably high proportion of missing cortisol data for which imputation was not a statistically sound option due to sensitivity analyses indicating that the data were not missing at random. The relatively short duration of the 12-week weight loss program may have attenuated the likely outcomes that could have been achieved with a longer intervention; however, weight losses achieved by the Lifestyle + Stress group were comparable with those seen among African American women in longer, more intensive studies (Boltri et al., 2008; Stolley et al., 2008; West et al., 2007), suggesting that the impact of combining stress management methods with weight control strategies may have a marked impact even in relatively brief programs. Finally, we did not systematically collect data to assess whether or not participants actually used the stress management techniques that were taught in the Lifestyle + Stress arm of the study. Future studies should include process data on the uptake of stress management strategies to strengthen the evidence for the efficacy of using stress management techniques to improve weight loss outcomes among a sample of moderate to highly stressed African American women. Despite being underpowered, this pilot, feasibility study offers promising trends that merit continued exploration.

In summary, this study demonstrates the feasibility of delivering a stress management–augmented behavioral weight loss program and its potential to improve weight loss outcomes in a sample of overweight/obese African American women with moderate to high perceived stress. The implications of our findings suggest that addition of a multifaceted stress management component to behavioral weight loss programs may be a viable method of improving weight loss outcomes in African American women, who typically achieve suboptimal weight loss during behavioral obesity treatment. Additional research is warranted to further examine the benefits of incorporating stress reduction techniques in behavioral weight control to improve weight loss outcomes and explore the relationship between stress reduction and weight loss success among overweight and obese African American women with high levels of perceived stress.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Ard

J. D.

Cox

T. L.

Zunker

Wingo

B. C.

Jefferson

W. K.

Brakhage

(2010). A study of a culturally enhanced EatRight dietary intervention in a predominately African American workplace. Journal of Public Health Management Practice, 16(6), E1-E8.

Ard

J. D.

Kumanyika

S. K.

Stevens

V. J.

Vollmer

W. M.

Samuel-Hodge

C. D.

Kennedy

B. M.

. . . Svetkey

L. P.

(2008). Effect of group racial composition on weight loss in African Americans. Obesity, 16, 306-310.

Banks-Wallace

(2007). Outcomes from Walk the Talk: A nursing intervention for Black women. Association of Black Nursing Faculty Journal, 18(1), 19-24.

Bjorntorp

(2001). Do stress reactions cause abdominal obesity and comorbidities? Obesity Reviews, 2(2), 73-86.

Bjorntorp

Rosmond

(2000). Obesity and cortisol. Nutrition, 16, 924-936.

Boltri

J. M.

Davis-Smith

Y. M.

Seale

J. P.

Shellenberger

Okosun

I. S.

Cornelius

M. E.

(2008). Diabetes prevention in a faith-based setting: Results of translation research. Journal of Public Health Management Practice, 14(1), 29-32.

Cohen

Kamarck

Mermelstein

(1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385-396.

Cole

S. R.

(1999). Assessment of differential item functioning in the Perceived Stress Scale–10. Journal of Epidemiology and Community Health, 53, 319-320.

Cox

T. L.

Zunker

Wingo

B. C.

Jefferson

W. K.

Ard

J. D.

(2011). Stressful life events and behavior change: A qualitative examination of African American women’s participation in a weight loss program. Qualitative Report, 16, 622-634.

10.

Delahanty

L. M.

Meigs

J. B.

Hayden

Williamson

D. A.

Nathan

D. M.

(2002). Psychological and behavioral correlates of baseline BMI in the diabetes prevention program (DPP). Diabetes Care, 25, 1992-1998.

11.

Diabetes Prevention Program Research Group. (2002). The Diabetes Prevention Program: Descriptive of lifestyle intervention. Diabetes Care, 25, 2165-2171.

12.

Economos

C. D.

Hildebrandt

M. L.

Hyatt

R. R.

(2008). College freshman stress and weight change: Differences by gender. American Journal of Behavioral Health, 32(1), 16-25.

13.

Farag

N. H.

Moore

W. E.

Lovallo

W. R.

Mills

P. J.

Khandrika

Eichner

J. E.

(2008). Hypothalamic-pituitary-adrenal axis function: Relative contributions of perceived stress and obesity in women. Journal of Women’s Health, 17, 1647-1655.

14.

Fitzgibbon

M. L.

Stolley

M. R.

Ganschow

Schiffer

Wells

Simon

Dyer

(2005). Results of a faith-based weight loss intervention for black women. Journal of the National Medical Association, 97, 1393-1402.

15.

Flegal

K. M.

Carroll

M. D.

Ogden

C. L.

Curtin

L. R.

(2010). Prevalence and trends in obesity among U.S. adults, 1999-2008. Journal of the American Medical Association, 303, 235-241.

16.

Fowler-Brown

A. G.

Bennett

G. G.

Goodman

M. S.

Wee

C. C.

Corbie-Smith

G. M.

James

S. A.

(2009). Psychosocial stress and 13-year BMI change among Blacks: The Pitt County Study. Obesity, 17, 2106-2109.

17.

Geronimus

A. T.

Hicken

Keene

Bound

(2006). “Weathering” and age patterns of allostatic load scores among Blacks and Whites in the United States. American Journal of Public Health, 96, 826-833.

18.

Hermann-Engels

Gretebeck

R. J.

Gretebeck

Jimenez

(2005). Promoting healthful diets and exercise: Efficacy of a 12-week after-school program in urban African Americans. Journal of the American Dietetics Association, 105, 455-459.

19.

Hollis

J. F.

Guillion

C. M.

Stevens

V. J.

Brantley

P. J.

Appel

L. J.

Ard

J. D.

. . . Weight Loss Maintenance Trial Research Group. (2008). Weight loss during the intensive intervention phase of the Weight-Loss Maintenance Trial. American Journal of Preventive Medicine, 35, 118-126.

20.

Kaiser

K. A.

Affuso

Beasley

T. M.

Allison

D. B.

(2011). Getting carried away: A note showing baseline observation carried forward (BOCF) results can be calculated from published complete-cases results. International Journal of Obesity. Advance online publication. doi:10.1038/ijo.2011.25

21.

Kanders

B. S.

Ullman-Joy

Foreyt

J. P.

Heymsfield

S. B.

Heber

Elashoff

R. M.

. . . Blackburn

G. L.

(1994). The Black American Lifestyle Intervention (BALI): The design of a weight loss program for working-class African-American women. Journal of the American Dietetic Association, 94, 310-313.

22.

Karanja

Stevens

V. J.

Hollis

J. F.

Kumanyika

S. K.

(2002). Steps to soulful living (steps): A weight loss program for African-American women. Ethnicity and Disease, 12, 316-319.

23.

Katzer

Bradshaw

A. J.

Horwath

C. C.

Gray

A. R.

O’Brien

Joyce

(2008). Evaluation of a “nondieting” stress reduction program for overweight women: A randomized trial. Journal of Health Promotion, 22, 264-274.

24.

Kennedy

B. M.

Champagne

Ryan

D. H.

Newton

Conish

B. K.

Harsha

D. W.

. . . Lower Mississippi Delta Nutrition Intervention Research Initiative. (2009). The “Rolling Store”: An economical and environmental approach to the prevention of weight gain in African American women. Ethnicity and Disease, 19, 7-12.

25.

Kim

K. H.

Bursac

DiLillo

V. G.

White

D. B.

West

D. S.

(2009). Stress, race, and body weight. Health Psychology, 28, 131-135.

26.

Kumanyika

S. K.

(1994). Obesity in minority populations: An epidemiologic assessment. Obesity Research, 2, 166-182.

27.

Kumanyika

S. K.

Espeland

M. A.

Bahnson

J. L.

Bottom

J. B.

Charleston

J. B.

Folmar

. . . TONE Cooperative Research Group. (2002). Ethnic comparison of weight loss in the Trial of Nonpharmacologic Interventions in the Elderly. Obesity Research, 10, 96-106.

28.

Kumanyika

S. K.

Gary

T. L.

Lancasster

K. J.

Samuel-Hodge

Banks-Wallace

Beech

B. M.

. . . Whitt-Glover

M. C.

(2005). Achieving healthy weight in African-American communities: Research perspectives and priorities. Obesity Research, 13, 2037-2047.

29.

Kumanyika

S. K.

Wadden

T. A.

Shults

Fassbender

J. E.

Brown

S. D.

Bowman

M. A.

. . . Wu

(2009). Trial of family and friend support for weight loss in African American adults. Archives of Internal Medicine, 169, 1795-1804.

30.

Look AHEAD Research Group. (2006). The Look AHEAD Study: A description of the lifestyle intervention and the evidence supporting it. Obesity, 14, 737-752.

31.

Luecken

L. J.

Gallo

L. C.

(2007). Handbook of physiological research methods in health psychology. Thousand Oaks, CA: Sage.

32.

Manzoni

G. M.

Pagnini

Gorini

Preziosa

Castelnuovo

Molinari

Riva

(2009). Can relaxation training reduce emotional eating in women with obesity? An exploratory study with 3 months of follow-up. Journal of the American Dietetic Association, 109, 1427-1432.

33.

McTigue

Larson

J. C.

Valoski

Burke

Kotchen

Lewis

C. E.

. . . Kuller

(2006). Mortality and cardiac and vascular outcomes in extremely obese women. Journal of the American Medical Association, 296, 79-86.

34.

Must

Spadano

Coakley

E. H.

Field

A. E.

Colditz

Dietz

W. H.

(1999). The disease burden associated with overweight and obesity. Journal of the American Medical Association, 282, 1523-1529.

35.

Ogden

C. L.

Carroll

M. D.

Curtin

L. R.

McDowell

M. A.

Tabak

C. J.

Flegal

K. M.

(2006). Prevalence of overweight and obesity in the United States, 1999-2004. Journal of the American Medical Association, 295, 1549-1555.

36.

Samuel-Hodge

C. D.

Johnston

L. F.

Gizlice

Garcia

B. A.

Lindsley

S. C.

Bramble

K. P.

. . . Keyserling

T. C.

(2009). Randomized trial of a behavioral weight loss intervention for low-income women: The Weight Wise Program. Obesity, 17, 1891-1899.

37.

Schiffman

S. S.

Graham

B. G.

Sattely-Miller

E. A.

Peterson-Dancy

(2000). Elevated and sustained desire for sweet taste in African-Americans: A potential factor in the development of obesity. Nutrition, 16, 886-893.

38.

Schulz

Israel

Williams

Parker

Becker

James

(2000). Social inequalities, stressors and self reported health status among African American and White women in the Detroit metropolitan area. Social Science & Medicine, 51, 1639-1653.

39.

Shao

Jordan

D. C.

Pritchett

Y. L.

(2009). Baseline observation carry forward: Reasoning, properties, and practical issues. Journal of Biopharmaceutical Statistics, 19, 672-684.

40.

Sharp

L. K.

Kimmel

L. G.

Kee

Saltoun

Chih-Hung

(2007). Assessing the Perceived Stress Scale for African American adults with asthma and low literacy. Journal of Asthma, 44, 311-316.

41.

Smith

D. W.

McFall

S. L.

(2005). The relationship of diet and exercise for weight control and the quality of life gap associated with diabetes. Journal of Psychosomatic Research, 59, 385-392.

42.

Stevens

(2000). Obesity and mortality in African-Americans. Nutrition Reviews, 58, 346-358.

43.

Stolley

M. R.

Fitzgibbon

M. L.

Schiffer

Sharp

L. K.

Singh

Van Horn

Dyer

(2008). Obesity Reduction Black Intervention Trial (ORBIT): Six-month results. Obesity, 17, 100-106.

44.

Strickland

O. L.

Giger

J. N.

Nelson

M. A.

Davis

C. M.

(2007). The relationships among stress, coping, social support, and weight class in premenopausal African American women at risk for coronary heart disease. Journal of Cardiovascular Nursing, 22, 272-278.

45.

Torres

S. J.

Nowson

C. A.

(2007). Relationship between stress, eating behavior, and obesity. Nutrition, 23, 887-894.

46.

Turner

R. J.

Avison

W. R.

(2003). Status variations in stress exposure: Implications for the interpretation of research on race, socioeconomic status, and gender. Journal of Health and Social Behavior, 44, 488-505.

47.

Vicennati

Pasqui

Cavazza

Pagotto

Pasquali

(2009). Stress-related development of obesity and cortisol in women. Obesity, 17, 1678-1683.

48.

Wadden

T. A.

Butryn

M. L.

Sarwer

D. B.

Fabricatore

A. N.

Crerand

C. E.

Lipschutz

P. E.

. . . Williams

N. N.

(2006). Comparison of psychosocial status in treatment-seeking women with class III vs. class I-II obesity. Surgery for Obesity and Related Disease, 2, 138-145.

49.

Walcott-McQuigg

J. A.

(1995). The relationship between stress and weight-control behavior in African-American women. Journal of the National Medical Association, 87, 427-432.

50.

Wang

Beydoun

M. A.

(2007). The obesity epidemic in the United States-gender, age, socioeconomic, racial/ethnic, and geographic characteristics: A systematic review and meta-regression analysis. Epidemiologic Reviews, 29, 6-28.

51.

Ware

J. H.

(2003). Interpreting incomplete data in studies of diet and weight loss. New England Journal of Medicine, 348, 2136-2137.

52.

West

D. S.

DiLillo

V. G.

Bursac

Gore

S. A.

Greene

P. B.

(2007). Motivation interviewing improves weight loss in women with type 2 diabetes. Diabetes Care, 30, 1081-1087.

53.

West

D. S.

Prewitt

T. E.

Bursac

Felix

H. C.

(2008). Weight loss of Black, White, and Hispanic men and women in the Diabetes Prevention Program. Obesity, 16, 1413-1420.